Between the restoration of the College Entrance Exam (Gaokao) in 1978 and the expansion of higher education admissions in 1999 (hereafter ‘The Expansion’), the scale of Chinese higher education remained relatively stable with slight growth. During this period, not only was admission to higher education extremely limited, but the structure of higher education was also obviously imbalanced between genders. Although the proportion of female students constantly increased over the years from 1978 to 1999, the figure never reached 40% (see Figure 1). However, since the implementation of The Expansion in 1999, the situation has changed radically. Figure 1 shows that the gender structure has undergone a significant change, with female enrollment at around 50% in 2010. At the same time, the opportunity gap in receiving higher education between men and women also decreased to such a degree that it can be disregarded (Hao [2010], p. 59; Liu [2006]). In this sense, women were the primary beneficiaries of The Expansion (Jin [2006]; An [2002]; Yang [2009]). The substantial and sustained increase of resources as a whole provided more opportunities for women, thus making gender equality in the field of higher education a reality.
However, sociological insights require us to do more than observe the absolute scale of higher education and the change of gender structure after The Expansion, or establish a direct causal relationship between The Expansion and gender equality in enrollments simply because of the conditions aforementioned. The Expansion and gender equality in educational opportunity are two mutually independent processes; the former does not necessarily lead to the latter (Mare [1980]; Wu [2009]). In theory, without specific institutional bias, the new educational opportunities from the expanding scale of higher education are not necessarily beneficial to the female population. In other words, the link between The Expansion and gender equality in opportunity still calls for further investigation, namely how The Expansion leads to this equality. Specifically, how are the new educational opportunities, which are leading to a higher degree of gender equality, allocated between genders after The Expansion? What are the factors that contribute most to the gender equality process?
Gender equality issues in higher education depend directly on the allocation of educational opportunities. Previous studies on higher education opportunity focus on three important factors. The first is the expansion of the scale of higher education. This increases the total amount of educational resources, bringing more opportunities to be distributed and thus changing the existing distribution structure and equality of status. Second is family background, especially families' socioeconomic conditions and parental educational level, which are considered to be the most crucial factors that restrict access to higher education. Third is the factor with the strongest Chinese characteristics, the household registration system. The policies of ‘Different Ruling Systems in Cities and the Countryside’ and ‘One Country, Two Policies’ contain in themselves or directly influence the unequal distribution of educational opportunities. Many research works in sociology and sociology of education have focused on these three factors and their relation to the equality of educational opportunity. The findings have formed three fields of discussion on higher education opportunity, and many studies fall into these three frameworks.
The expansion of education and educational opportunity
Whether in industrialized countries or those still in progress, more and better education determines whether people rise to the top of the social hierarchy (Deng and Treiman [1997]). In a study on status attainment patterns and intergenerational mobility, Blau and Duncan ([1967]) used empirical data to test this correlation and confirmed that education significantly influences both an individual's initial and present occupation. Chinese scholars used China's local data to refit the model and found the impact of education on initial jobs to be strikingly consistent in both China and the USA (Bian et al. [2006]). In other words, higher education largely determines the socioeconomic status of individuals and improves social structure as well. Therefore, after World War II, many countries implemented expanding policies to increase their citizens' opportunities to receive higher education. The policies' design can popularize higher education and, to a certain extent, decrease the effect of external factors, such as socioeconomic conditions, on restraining access to higher education while expanding the impact of personal endeavor, talent, and so on.
Nevertheless, these policies' intended effects are rarely supported by empirical evidence; the reality turns out to be precisely the opposite. In many industrialized countries, with the expansion of education, the impact of family background on educational opportunity is not reduced but remains at the original pace. In a study comparing 13 countries, Shavit and Blossfeld ([1993]) found that in most countries, despite the effort put into expanding higher education, equality of educational opportunity was not promoted but in fact the gap in enrollments between the dominant and nondominant class widened. Many Chinese scholars also found that the process of expansion expanded the dominant class' comparative advantage in higher education (Liu [2006]). For example, Yu Li's study found that the managing stratum increased their advantages while the advantages of other nonlabor classes remained unchanged (Li [2006]). Xiaogang Wu's studies showed that in 2000, the relative educational opportunity for disadvantaged groups was less than a decade ago (Wu [2009]). In this sense, the expansion of higher education did not have an obvious impact on achieving equality as expected.
The above studies reveal a special relationship between the expansion of education and educational opportunity. They also reflect the paradox between the intention of policies and the actual results. This phenomenon has been dubbed the ‘Maximally Maintained Inequality Hypothesis’ (abbreviated as ‘MMI Hypothesis’), indicating that an increase in educational opportunity will always benefit the dominant class first rather than the nondominant class; only after the fulfillment of the dominant class' educational needs at a particular stage will educational opportunities flow to the nondominant class (Raftery and Hout [1993]). Using the MMI Hypothesis, Chinese scholars have focused mainly on explaining the distribution of opportunity when investigating the impact of The Expansion, e.g., (Liu [2004], [2005], [2006]; Li [2006]; Wu [2009]; Hao [2010]). Dahai Hao used data from CGSS2003 to examine the educational conversion ratio at various educational stages between 1949 and 2003. The results showed that China already reflects the characteristics of the MMI Hypothesis: in high school and college, the dominant class has a stable advantage, while compensatory policy targeting the middle and lower classes has not achieved satisfactory results (Hao [2010], p. 64). Using the same data, Yu Li found that since 1992, the effect of family educational background on years of education is on the decline, but educational inequality due to class division is on the rise. This is in line with the MMI Hypothesis and even exceeds its expectations (Li [2006]). Jingming Liu's study found that the influence of parental class on their children's schooling opportunities showed an opposite change: in survival-level education, the offspring of the dominant class experienced a clear decrease in comparative advantage, while for education leading to a rise in status, the relative advantage of the dominant class displayed an upward trend (Liu [2006]).
Family background and educational opportunity
The family incorporates economic, cultural, and social capital in one unit and controls for the total capital. Its ability and intention to invest in education significantly influences the children's educational opportunities. Specifically, families from the dominant class can use their dominant position to gain more educational opportunities and are also able to make use of their own capital to compete for various types of high-quality educational resources. Thus, the hierarchy of capital advantage can transform into educational advantage for the next generation. For those families from the nondominant class, many obstacles to educational opportunities exist. Ascribed family background can thus be transmitted across generations. Bourdieu found that the sons of senior staff have a much better opportunity to enter college compared with those of agricultural and skilled workers (Bourdieu and Passelon [2002], p. 6). A more specific study in China show that if the father's monthly income is higher than 2,000 RMB (about 330 USD), his children will have a better chance of entering a university or college for further study (Li [2010]). Specifically, two paths are shown to transfer family background into educational opportunity: the ‘Resource Conversion Mode’ and the ‘Cultural Reproduction Mode’ (Li [2006]).
The Resource Conversion Mode refers to the point at which, while transferring the family's social economic capital into educational opportunities for their children, two exclusion mechanisms (direct exclusion and indirect exclusion) are imbedded in the process, leading to the intergenerational transmission of class inequality in the field of education. The first exclusion mechanism is the educational investment capacity of families. A strong economic background allows children from the higher class who are not able to fulfill admission requirements to access the next level of education or better educational environment by paying ‘sponsor fees’ or other means. Children from lower class backgrounds are thus excluded from formal education due to inability to pay tuition and sponsor fees (Li [2006]). A study of ten cities in China showed a strong disparity across different classes where educational cost occupies a distinctively different proportion of income: an urban manual worker can barely afford one child's college education, whereas a manager can afford that of three (Liu [2005], pp. 237-238). The second exclusion mechanism is the family's intention to provide further education for their children. In considering the investment and reward of education, some individuals may voluntarily withdraw from academic competition (Li [2006]). Willis found that in a boys' school in Britain, working-class boys tended to give up on study and voluntarily engage in manual work. Selling their labor in nontechnical work is an easy way to support their families (Hao [2010], p. 42). In the USA, a tuition fluctuation of 10% will affect 6.2% of students in deciding whether to continue their education by entering college (Cheng [2002], p. 123).
The Cultural Reproduction Mode refers to the influence of families' cultural capital on their children's learning motivation, interest, and performance at school, all of which determines the ability to acquire educational opportunity (Li [2006]). People from different cultural backgrounds inherit different quantities and types of cultural capital. Just as with economic capital, it is able to create, nurture, and be passed on to the next generation and has a cumulative impact on education as well as socialization, thus affecting the obtaining of socioeconomic status and opportunities for upward mobility (Li and Lv [2008], pp. 193-194). As the Coleman Report and the Plowden Report pointed out, ‘it is not the quality of the school but the child's social background that has a more significant impact on their academic achievement’ (Liu [2005], p. 43). Many scholars empirically confirm this transformation by measuring specific social capital indicators of families (Blau and Duncan [1967]; Hao [2010]).
Urban-rural differences and educational opportunity
The split between urban and rural areas due to the urban and rural household registration systems profoundly impacts the inequality in China. The urban-rural gap in higher education opportunity is an important manifestation of this inequality. This gap has drawn great concern in both politics and academia. Numerous studies (Zhao [2000]; Yang [2006]; Wang [2011]; Liu [2005]; Qiao [2008]; Li [2010]) have found that since The Reform and Opening Up, whether after the restoration of the College Entrance Exam or The Expansion in 1999, the gap in higher education opportunity between urban and rural areas did not improve and even deteriorated. For example, the declining proportion of rural students seriously worsened (Cao [2012]). Using 1% of the 2005 census data, Chunling Li found that the urban population born in the period 1975 to 1979 had a 3.4 times higher chance of getting into college compared to its rural counterpart, while the urban population born in the period 1980 to 1985 had a 5.5 times higher chance (Li [2010]). Another longitudinal study on educational inequality trends between 1981 and 2006 found that inequality between urban and rural areas in educational opportunity increased by 33.6% after The Expansion (Guo [2008]). More importantly, this urban-rural gap is not a regional but a national trend across China. Even in developed regions like Shanghai, Beijing, and Jiangsu, the gap between urban and rural years of education is approximately 2 years, while in underdeveloped regions like Guangxi, Guizhou, and Shaanxi, the gap was between 3 and 4 years of difference. In higher education, this difference in years of education results in an extremely disproportionate ratio between urban and rural college students in the total population. Data show that 8.5% of the total urban population has a college degree or above, while in rural areas, this proportion is only 0.6% (Zhang and Wu [2008], p.88).
Another trend in inequality of opportunities is the more significant urban-rural gap in high-quality higher education. Inequality in status-oriented higher education (mainly college and graduate studies) exhibits a wider gap than survival-based higher education (mainly adult education and junior college education). The coefficient of the former is 0.752, while that of the latter is 0.566 (Liu [2005], p. 269). Chunling Li further found that the urban-rural gap in college education is greater than in junior college education. Moreover, after The Expansion, this gap widened more in college education than in junior college education. The opportunity for urban residents to receive junior college education is 4.9 times higher than rural residents and 6.3 times higher for college education (Li [2010]). In order to get the same quality of education, students from rural areas need to make much more effort than urban students (e.g., spending one or several more years in high school) (Wang [2011]). Generally speaking, the higher the quality of educational resources, the larger the urban-rural gap in distribution of opportunity. Data show that the proportion of rural students enrolled in key national universities has reached a record low. In recent years, the proportion of rural students admitted into Peking University, Tsinghua University, and other prestigious universities is less than 20%; there has clearly been a serious setback compared with the 1980s (Cao [2012]; Qiao [2008]).
Continuous study of the three factors that influence higher education opportunity laid the foundation for further investigation into the inequality in higher education in China. However, simply applying the above findings to interpreting the distribution of higher education opportunity between genders and its equalization process produces a series of new questions, such as why didn't the advantaged and disadvantaged status of men and women prior to The Expansion remain unchanged, and whether gender differences exist in the Resource Conversion Mode and Cultural Reproduction Mode. If differences do exist, what are their specific manifestations? Did they change over the course of The Expansion? Also, does the urban-rural gap have different implications for the two genders? These new questions do not imply that previous studies on higher education opportunity neglected the issue of gender. In fact, issues related to gender differences have always been one of the important concerns in almost every research work (Liu [2004]; Li [2006]; Zheng and Li [2009]; Wu [2009]; Hao [2010]; Lu [2004]; Li [2009]; Ye and Wu [2011]; Wu [2012]), but most of the studies treated gender analytically and methodologically as an independent variable in the distribution process. Thus, gender distribution is often presented as an independent aspect of educational opportunity distribution, and these works lack interactive analysis between gender and factors such as The Expansion, family background, and urban and rural differences.b That is, there is a lack of specialized study on gender distribution of higher education opportunity that incorporates a variety of dimensions.
Treating the gender gap as an independent aspect of the distribution of educational opportunity is obviously inadequate. Gender affects the aforementioned factors, and men and women are influenced by these factors differently (Li [2009]; Wu [2012]; Ye and Wu [2011]). This requires further investigation on how different factors influence gender distribution respectively as well as how they influenced the equalization process after The Expansion. By looking at the interactive effect between gender and other factorial variables, this study investigates whether factors that influence the distribution of educational opportunities have a significantly different impact on men and women. In this study, gender interacts specifically with family economic conditions, parental educational level, household registration type, and other factors and compares how educational opportunities distribute between genders prior to and after The Expansionc. This comparison identifies the factors that contribute most to the gender equality process.
The first interaction term is family economic conditions and gender (i.e., family economic conditions × gender). This interaction term analyzes how family economic conditions influence gender distribution of educational opportunities differently between men and women. Numerous studies have demonstrated the positive correlation between family economic conditions and higher education opportunity (Liu [2000], [2004], [2005], [2006]; Li [2006]; Wu [2009]; Ye and Wu [2011]; Li [2005], [2010]). Better economic conditions can lead to greater ability to support children's education and gaining greater opportunity in higher education for the children; if the converse situation, the opportunity is smaller. After the restoration of the College Entrance Exam in 1977, women's opportunities to access higher education have long been lower than those of men. Interacting gender with family economic conditions may introduce new influence from economic disparity or expand the existing inequality. Better family economic condition can decrease the number of students forced to give up further studies due to the cost of education. This is undoubtedly beneficial for girls, who are traditionally disadvantaged in this distribution. When the economic conditions become restrictive, the patriarchal tradition tends to sacrifice girls' opportunities for higher education, leading to their early entry into the labor market. Thus, Hypothesis 1: The gender gap in higher education opportunity negatively correlates with family economic conditions. In particular, groups from families with better economic conditions face a narrower gender gap than those from families with poorer economic conditions.
The second interaction term is parental educational level and gender (i.e., parental educational level × gender). This interaction term analyzes how parental educational level influences gender distribution of educational opportunities differently between men and women. The Cultural Reproduction Mode has noted that the higher the educational level of parents, the greater the opportunity for children to receive higher education, and if the converse, the opportunity is lower. Introducing gender into the question can investigate whether parents' educational level has a different impact on men and women in accessing higher education. Typically, the higher the parents' education level, the more likely they will accept the ideology of gender equality and the gender gap in educational investment will be smaller; the lower the parents' educational level, the more likely they will accept the patriarchal concept, which will reduce women's opportunity in higher education. Thus, Hypothesis 2: The gender gap in higher education opportunity negatively correlates with parents' educational level; in particular, people with parents who have a higher educational level have a smaller gender gap than those with parents who have a lower educational level.
The third interaction term is differences between urban and rural and gender (i.e., urban-rural differences × gender). The purpose of creating this interaction term is to analyze how urban-rural differences influence gender distribution of educational opportunities differently between men and women. According to the aforementioned discussions on urban-rural differences and educational opportunity, higher education opportunity differs drastically between urban and rural areas. Adding gender into the picture will likely produce the following characteristics: rural households are disadvantaged in economic conditions, gender ideology, and educational level, and the number of children is larger than in urban households. This will obviously restrain women's access to higher education, and thus, the gender gap in higher education opportunity is relatively large. In contrast, for urban households, economic conditions, gender ideology, educational level, and the number of children will not necessarily constrain female access to higher education; thus, the gender gap in higher education opportunity is relatively small. In other words, gender inequality is more serious in rural areas. Thus, Hypothesis 3: The gender gap in higher education opportunity is narrower in urban areas than in rural areas.
The three interaction terms and the corresponding research hypotheses are only the foundation for investigating The Expansion and gender equality in higher education opportunity. In order to further investigate how the new educational opportunities are allocated between genders, The Expansion needs to be introduced as a variable as well. To be specific, three interaction terms need to be viewed separately prior to and after The Expansion in different regression models. By comparing them, it will be possible to observe whether the impacts from the three interaction terms changed due to The Expansion. In other words, what changed after The Expansion in the influence of family economic conditions, parents' educational level, and urban-rural differences on gender distribution of higher education opportunity? According to hypotheses 1, 2, and 3, the gender gap in higher education is already relatively small for advantaged groups; it is more important to track the change in disadvantaged groups. Thus, it is more important to observe the gender gap change in groups with lower economic conditions, lower parental educational level, and living in rural areas prior to and after The Expansion, and determine whether it is the fact that women in these groups become the beneficiaries that contributes to the gender equalization process in higher education. We therefore propose the following:
Hypothesis 4: After The Expansion, the gender gap in higher education opportunity narrowed more in groups with lower family economic conditions.
Hypothesis 5: After The Expansion, the gender gap in higher education opportunity narrowed more in groups with lower parents' educational level.
Hypothesis 6: After The Expansion, the gender gap in higher education opportunity narrowed more in rural areas than urban areas.