Elite Mobility and Conversions of Different Forms of Capital: An Investigation of Patterns of Study Abroad amongst Elite Graduates from Peking University in China

ABSTRACT This article extends Bourdieu’s convertibility of different forms of capital to understand the patterns of study abroad by elite graduates from Peking University, China. We draw upon empirical data from a first-hand survey study involving 1,417 graduates from Peking University. The statistical analyses suggest a pattern of the conversions from political and economic capital to cultural capital. Students from leading cadres and managerial families up the prestige game by actively pursuing study abroad opportunities, thus creating a ‘hard currency’ which combines elite status at home and abroad. This prestige game locks out even elite graduates without adequate economic resources. The formation of the new ‘hard currency’ reveals the inherent inconsistencies between the State’s meritocracy discourse and its ambition of competing for global talents. We argue that the State’s recent pursuit of global talents effectively excludes the working class and agricultural families without providing an inclusive and convincing meritocratic rationale.


Introduction
The mobility of international students has increased dramatically during the era of globalised higher education systems (Courtois 2018) (Brook and Waters 2010;Waters 2005). Chinese international students have attracted considerable research attention as they have become the largest group amongst those who pursue their studies abroad (Bodycott 2009;Kajanus 2016;The Economist 2018). Prior research highlights the rise of Chinese urban middle class and their increasing spending power as an important drive for the study abroad market (Shen 2018). Many studies have focused on individual-level variables, such as Chinese students' choices of country destinations (Bodycott 2009) or the patterns of choices in universities and fields of study (Wu 2014;Cebolla-Boado, Hu, and Soysal 2018) or the cultural adaptations in overseas institutions (Zhang and Xu 2020;Gu and Schweisfurth 2006). Some studies have further explored the role of the state in pursuing global leadership in research and innovation, which has brought upon significant implications on the international outlook of university education and degrees (Liu 2016;Lan 2019).
These studies have enriched our knowledge about Chinese international students and their study abroad patterns; however, some obvious gaps exist. Generally, Chinese international students are more likely from affluent families (Henze and Zhu 2012;Wei 2013;Quan, He, and Sloan 2016;Cebolla-Boado, Hu, and Soysal 2018). For instance, an estimate of 544,500 students left China in 2016 to study abroad, and 91.49% of this cohort were financially sponsored by themselves or their families (Lan 2019). What is less known is the implications of students' academic backgrounds or university origin in China on their study aboard chances. For instance, Shen (2018) finds that around 13.78% of graduates from elite universities chose to further their postgraduate study abroad, but only 2.01% of non-elite graduates from public universities did the same. The two lines of inquiries on the social and academic origin of Chinese international students have rarely been examined together in an empirical study. The present study fills in these gaps by focusing on a case study of one elite university and using original surveys to map out how social origin and demographic factors shape elite Chinese students' choices of study abroad.

Bourdieu's forms and convertibility of capital
Boudieu's cultural capital theory highlights three types of capital, namely, economic, cultural and political capital, in 'materialised' or 'embodied' forms, which shape social stratification in a society (1986). More importantly, a fluidity of conversions amongst different types of capital, from liquidity to economic capital, from cultural capital to economic capital and from social capital to economic capital (Bourdieu 1986, 243). Bourdieu (1986:, 243) then argues that these transactions can be further institutionalised in 'property rights', 'educational qualifications' and 'social titles'.
Given that the conversions from different types of capital are transactions, they also involve risks, uncertainty and delays of returns (Bourdieu 1986). For example, the conversion from economic capital to cultural capital not only takes time-from childhood to university education-but it is also costly in terms of 'the purchase of the time of others', such as mothers' time spent in childrearing and child care (Bourdieu 1986, 253). When comparing different risk levels of these transactions (Bourdieu 1986), cultural capital is perhaps the costliest but most disguised investment that will reproduce and warrant privileges and wealth. As Bourdieu (1986:, 255) puts it, 'as an instrument of reproduction capable of disguising its own function, the scope of the education system tends to increase. And together with this increase is the unification of the market in social qualifications, which gives rights to occupy rare positions'.
Bourdieu explains the complexity of different types of privilege, but more importantly, how the elites can afford to make 'transactions' amongst different types of capital legitimate. For instance, studies on elite universities highlight that these 'rare positions' in the educational system are often obtained by students from 'rare positions' in society, who share similar habitus, cultural cultivations and privileged socioeconomic backgrounds (Liang and Lee 2012;Power et al. 2013;Binder, Davis, and Bloom 2016;Brown et al. 2016). However, elite reproduction through the education system is not simply a 'scholastic mode of reproduction' (Bourdieu andPasseron 1977 [1990]) but also requires a constant adaptation and readjustment by the elites to changes in the social structure particularly in the era of degree inflation and overqualification (Green 2017). Brown et al.'s (2016) recent study highlights the narrative shift from elites to 'talents' amongst elite graduates in France and the UK. This shift is explained by the increasing level of competition for elite opportunities in society as the middle-class and higher education system have expanded (Power et al. 2013;Brown et al. 2016).
Moving beyond reproducing elite students' individual dispositions and habitus through elite education, some scholars have highlighted macrolevel factors that complement persistent elite reproduction. For instance, campus mechanisms, such as on-campus recruitment and alumni networks, play an important role in defining occupational prestige and creating an aspiration hierarchy that prioritises high-tech and investment firms in the US context (Ho 2009;Binder, Davis, and Bloom 2016). Power et al.'s (2013) comparative study between Oxford and Sciences-Po graduates shows how cultural characteristics at the contextual level continue to shape elitism. A pronounced sense of self-fulfilment and expressive individualism exist amongst Oxford graduates in contrast to a pattern of elite career aspirations inspired by the French civic tradition and a culture of nationhood from the perspectives of Sciences-Po graduates (Power et al. 2013).
The complexity of the conversions between different types of capital and fluid conversions into cultural capital obscure social reproduction processes, which not only conceal but also legitimise the power and privilege of the elites. These theoretical perspectives have rarely been extended to the Chinese context. Therefore, the present study will make a start by using a case study of Peking University (PKU) to examine the conversions of different capitals and elite formations through cultural capital in China.
Elite opportunity pathways and the making of the Chinese elites Many scholars from a range of disciplines have explored the formation of the contemporary elite class during China's transition from a socialist society to a market one; and there exist three main pathways to elite social status. Political scientists such as Walder, Isaacson, andLu (2015, 2013) have highlighted the importance of political capital associated with the membership of the Chinese Communist Party during China's transition. Their studies show that minimal political change gave the Communist Party elites premium access to new economic opportunities and that political capital allowed some of them to transform into corporation or entrepreneurial elites during the market reform (Walder 2014(Walder , 2003Walder, Isaacson, andLu 2015, 2013). Different from Walder's account, Nee and Opper's (2012) research highlights a 'bottom-up' formation of entrepreneurs; they establish economic independence by decoupling from the Party members and suppliers of state-owned enterprises. The key to their successful transformation is the buildup of self-organised industry clusters with well-functioning networks of suppliers and distributors (Nee and Opper 2012).
Finally, perhaps the most important one is educational pathways as the key to elite status (Wu 2017). These educational pathways have been carefully maintained by social policies such as the Household Registration System, the hukou, which not only legitimises the rural-urban divide but also limits access to schooling amongst rural migrant children (Wu and Treiman 2004;Wu 2017). For instance, Wu and Zhang's (2010) research highlight that the inequality in the provision of education and progression in primary and secondary schooling is closely correlated with the rural-urban division. This finding is further supported by Tam and Jiang's (2015) research on higher education opportunities. They argue that rural students are more likely to choose a less prestigious pathway through higher education, such as vocational colleges, and that their previous poor quality of schooling prevents them from making more ambitious choices in higher education (Tam and Jiang 2015).
These studies provide important insights into the formation of social elites during China's market transition. However, what has been less explored amongst these three strands of literature is the formation of elites through conversions of different capitals. For instance, Goodman (2014) calculates that nearly 90% of the millionaires in China are direct offspring of high-ranking Party members in a managerial position. By highlighting the conversions into different forms of capital across different generations at the macrolevel, he further traces how pre-Revolution elites were transformed into political elites during the Socialist regime and later into business or corporation elites during the market reform (Goodman 2014).
In addition to the convertibility between political and economic capitals, some sociologists tend to focus on the role of education as a reliable investment channel to reproduce the privileges of the elites (Hu and Wu 2019;Loyalka et al. 2017). Elite universities are more likely to recruit students from privileged backgrounds, despite seemingly meritocratic selection of the Gaokao (Liu 2016). For instance, Liang and Lee (2012) traces the family characteristics of those enrolled in PKU, the most prestigious university in China, from the 1980s to the early 2000s, a period prior to the emergence of mass higher education. Their analysis shows that students from managerial cadres and professional families accounted for 79.19% of the total student population between 1985 and 1994, in contrast with their corresponding proportion of 7.1% of the national population (Liang and Lee 2012). This elite participation did not seem to change from 1995 to 1999. Furthermore, the students from the managerial cadre backgrounds were 17 times more likely to be accepted by PKU than their counterparts from professional, working class and peasant backgrounds between 1985 and 1994; and their advantage increased to 23 times more likely to attend the prestigious institution between 1995 and 1999 (Liang and Lee 2012).
We have known that elite social groups have maintained their privileges by converting their 'elite power' into different forms of capital. Moreover, privileged social groups have better access to elite schooling and universities than their working-class counterparts. However, with the increasing number of Chinese students seeking higher education qualifications abroad, what are the patterns for social elites to up their game of consolidating their privileges through studying abroad? This study will make a case of the conversions amongst political, economic and cultural capitals in creating a new 'hard currency'-a combination of elite certificates at home and abroad by elite graduates in China. This study asks a number of questions, of which the principal ones are: (1) What are the patterns of study abroad amongst the graduates from PKU? (2) How are these patterns affected by one's socioeconomic, cultural and demographic backgrounds?

Data and methods
The main data for this study are drawn from a project commissioned by the Office of Academic Affairs and conducted by the authors' team at PKU. This project aims to establish the patterns of study abroad by graduates from PKU. It incorporates a survey study amongst 1,417 graduates in 2017. This survey aims to explore the patterns of graduates studying abroad in relation to their social backgrounds from PKU. The questionnaire was sent via PKU's official email addresses to all the graduating classes with a total number of 3,074 students between June 8 and July 30, 2017. This study was also promoted via popular social media applications, such as WeChat and Weibo, as well as via the official PKU website. The final number of valid questionnaires was 1,417 after excluding questionnaires with missing values for key variables. The final response rate was 46.10%. By using population data provided by PKU, a bias analysis was undertaken to consider the representation of sample. The sample was found to reflect the wider population well in regards of age, gender and subject and hukou. Amongst the questionnaires returned, male students accounted for 57.7%, whereas the total male students in PKU accounted for 57.8%. Rural students in the survey accounted for 16.7%, whereas its total population was 14.2% amongst all PKU undergraduates. The proportion of the surveyed students from the fields of natural sciences was 30.4% compared with 28.5% in all undergraduates. Those from information and engineering sciences represented 17.6% in the survey in comparison to 15.5% in the total population. The surveyed students from humanities, social sciences and economics and management accounted for 15.4%, 18.3% and 11.7%, whereas their correspondent population was 16.8%, 19.6% and 13.3%, respectively. Thus, our surveyed sample provides a reasonable representation of PKU undergraduates in terms of gender, geographical origin and fields of study.
The design of the questionnaire survey elicits information on independent and dependent variables to investigate how students from different social and cultural backgrounds, as well as those from different fields of study, make choices to study abroad. The former includes socioeconomic status, parental educational level, cultural capital, geographical origins, gender, academic capital, grade point average (GPA) and experiences of short-term study abroad. The latter includes choices of study abroad and choices of study abroad by fields of study.

General surveyed student population by socioeconomic characteristics
The main purpose for the survey study is to map out the relations between students' socioeconomic and demographic backgrounds and their choices of study abroad. Table 1 details the coding and mean of all the independent and dependent variables. The first independent variable used is socioeconomic status. Drawing from Lu's (2010) classification of the contemporary social class in China, this survey categorised four social groups: leading cadres and those with managerial backgrounds, those with professional backgrounds, the working class and agricultural workers' backgrounds. Parental educational level was coded at postgraduate degrees, undergraduate degrees, senior secondary schooling and junior secondary schooling or below. The coding of economic resources of one's family was based on students' self-reported answers, with 1=bottom quintile and 5=top quintile, with a list of income quintiles in China attached to the appendix in the questionnaire.
Academic performance was measured by GPA results, which is the average academic score for the total number of courses undertaken at PKU during the four years of undergraduate study. Geographical origin refers to hukou status, the Household Registration system that distinguishes agricultural from non-agricultural residency. The dependent variable, the choices of study abroad, was measured by whether the students received at least one offer from an overseas institution at the time of the survey. Table 2 demonstrates the patterns of socioeconomic and demographic characteristics of those with overseas offers and compares this population to the total surveyed student population at PKU and to that of the general population in China. Amongst the sampled population, around 43% of the graduates come from elite social status, such as leading cadres and senior executives, whereas only 6% of the graduates come from the total population. Another one-third of the surveyed PKU graduates come from professional backgrounds in contrast with 15.8% of professionals in the total population. By contrast, only 11% and 7% of students are from urban working class families and agricultural families, respectively. This representation of the students from working class and agricultural backgrounds contrasts with around 30% and 40% of each social group in the total population. Evidently, the students from privileged families, such as leading cadres, senior executives and urban professionals, accounted for the majority of students who were studying at PKU. This findings confirm the previous study by Liang and Lee (2012) on persistant patterns of elitism in access to Peking University.
The socioeconomic patterns of those with overseas offers and the surveyed population were further studied. In Table 2, working class students account for around 11.6% of the surveyed student population at PKU; however, only 5.9% of students who received overseas offers were from the same backgrounds. A similar pattern was observed in the changes of students from agricultural families. Students from agricultural families accounted for 7% of the surveyed population but only 1.4% had recieved offers from universities abroad. By contrast, a distinctive rise was observed in the proportion of students from leading cadres and managerial backgrounds in receiving overseas postgraduate offers compared with their proportion of the surveyed student population. They accounted for around 43% of the surveyed student population but increased to around 54.1% of all students who received overseas offers. A similar pattern is also observed among students from professional backgrounds. They represented around 26.9% of the surveyed population but 32.4% of those who were offered a place to study in a university abroad.
These simple statistical analyses confirm the general cultural capital thesis that students from privileged backgrounds are not only more likely to be enrolled in elite universities, such as PKU, but are also more likely to pursue postgraduate opportunities abroad. Moreover, students from urban working class and agricultural families are the minority in elite places, such as PKU. Furthermore, these students are less likely to convert their elite qualifications into offers of further study abroad than their counterparts from privileged families. The next section will explore the roles played by socioeconomic status, economic capital and cultural capital in determining students' overseas postgraduate opportunities.

Multivariate analysis results
A hypothesis concerning cultural capital developed from previous discussions is that studying abroad suggests a strong association between students' social origin and their chances of gaining overseas offers and that academic performance would have little influence on one's chances of gaining offers abroad, regardless of one's social origin. A series of hypotheses can be formulated regarding the influence of socioeconomic background, cultural capital and sociodemographic characteristics on the offers received by graduates from PKU. Moreover, socioeconomic and demographic backgrounds would be examined along with academic performance in relation to the chances of receiving overseas offers. Table 3 further reports the results from a series of logistic regression analyses of the log odds of the independent variables on studying abroad, where at least one offer received from an overseas institution was coded as 1, and no offer from abroad was coded as 0. The logistic regression predicts the log odds that an observation will have an indicator equal to 1. The odds of study abroad are defined as the ratio of the probability that a student receives at least one offer to the probability that the graduate does not receive any offer. Model 1 shows the net effect of socioeconomic status, economic capital and cultural capital on study abroad offers. In this model, students from professional and managerial backgrounds are 1.569 times more likely to 5.9 Sources: The data on socioeconomic participation in higher education are generated from a previous empirical work; the data on the composition of the corresponding population come from Lu (2010). Note: *The first two socioeconomic groups are combined together, that is, the leading cadres/governmental officials and executive personnel.
pursue further education opportunities abroad than those from working-class and agricultural families. Students from families with higher levels of economic resources seem to have significant advantages, that is, they are 66% more likely to study abroad than those from lower-income families. Both parents' levels of education are important predictors of gaining offers to study abroad. Models 2 and 3 introduce a series of sociodemographic factors, the hukou and gender, as additional predictors that may influence the chances of studying abroad. The effects of socioeconomic status and parental educational level are also controlled to capture the changes of the effects of social backgrounds. Thus, students from urban areas are significantly more likely to pursue overseas postgraduate opportunities than their counterparts from rural areas. When geographic origin is considered, the effect of socioeconomic origin declines. However, the effects of economic resources and parental levels of education are still significant. Opportunities to study abroad are not conditioned by gender.
Model 4 introduces a meritocratic measure, namely, GPA. It shows that the GPA measures are significant in predicting overseas opportunities. When the GPA is included, the effect of parental educational level reduces moderately, whereas the influence of economic capital is even more significant. Model 5 introduces the prior experience of short-term study abroad during undergraduate years. Hence, the students with experiences of prior study abroad are 2.174 times more likely to choose to pursue postgraduate opportunities abroad. Table 4 provides breakdown details on students' socioeconomic and demographic characteristics and their fields of study in relation to overseas offers. The survey covers six fields of study at the undergraduate level at PKU, including STEM subjects, such as the natural sciences, information technology and engineering, as well as non-STEM disciplines, such as social sciences, economics and business management, humanities and interdisciplinary studies. The largest proportion of the overseas offers came from natural sciences, with a total of 166. Another 82 graduates from information Df N 0.00408**** 6 1,367 0.00408***** 7 1,367 0.00349*** 8 1,367 0.00235*** 9 1,367 1.38e-05*** 10 1,367 Robust Standard errors in parentheses. ***p < 0.01, **p < 0.05, *p < 0.1 technology also received the offers to pursue postgraduate degrees abroad. Graduates from non-STEM disciplines accounted for 43.5% of the total of overseas offers (191 postgraduate offers). Table 4 further reports the results from logistic regression analyses of the log odds of independent variables on study abroad, where at least one offer was received from an overseas institution in relation to the fields of study. It shows the variations of the effects of economic capital, parental educational level, gender, GPA results and prior overseas experience on the likelihood of receiving overseas offers by fields of study. Economic capital is important in determining the overseas offers for graduates of STEMs and non-STEM subjects, except social sciences and interdisciplinary study, when all other factors are controlled. The mother's level of education seems to be more important than father's level of education in terms of determining the chances of receiving overseas postgraduate offers; and it is a significant factor for all graduates, except for graduates in the information technology and engineering and business management. Again, no significant difference exists between male and female students in receiving overseas offers, except for graduates from the natural sciences, information technology and engineering. GPA results and experience of prior study abroad seem to affect graduates of the STEM subjects more significantly than those of non-STEM fields.
These statistical analyses partially support the cultural capital and capital conversion thesis in the context of PKU graduates and their patterns of pursuing opportunities to study abroad. Firstly, students' socioeconomic status and cultural capital seem to play an important role in determining overseas study offers. Students from privileged urban backgrounds and those from educated families are considerably more likely to pursue postgraduate study opportunities abroad than their counterparts from working class and rural agricultural families. These social and cultural advantages might allow the elite graduates from privileged families to 'edit' the prestige hierarchy by furthering their competitiveness and extending their elite status abroad. These strategies seem to be unavailable for those from working class and agricultural families.
Secondly, these findings highlight the persistent significant influence of economic capital in determining study abroad offers, even when socioeconomic status and cultural capital are considered. Even when the meritocratic measure, GPA results, is included, the families' income level still makes a significant influence. When it comes to the breakdown details on fields of study, economic capital is still a consistent predictor of overseas study offers amongst most STEM and non-STEM subjects. This finding may suggest that the cost of studying abroad might have put off students from less affluent families. Thus, students from affluent urban families are the winners of the 'study abroad' game. This finding might suggest a tendency of convertibility from financial capital to elite cultural capital amongst PKU graduates in China.
Thirdly, cultural capital, measured by both parents' level of education, is significant in predicting overseas postgraduate offers when other socioeconomic and demographic factors are included. However, the mother's level of education seems to be more important than the father's level of education when specific fields of study are concerned. This finding is consistent with studies on parental reproduction of cultural capital in China (Sheng 2012). Arguably, mothers are more likely to spend more time in emotional and educational support of their children than fathers do, and mothers play an important role in raising the educational aspirations of their children (Sheng 2012).
Fourthly, the findings show that female students are as likely to study abroad as their male counterparts except those from the field of natural sciences. This result can be explained by the unique demographic characteristics of all the participants in the survey who were born into the one-child generation (Liu 2021). Prior studies have suggested that girls from the one-child cohorts have achieved representation equal to their male counterparts in higher education (Liu 2017). As urban families have only one child (Fong 2004), they are likely to invest in their only child's postgraduate opportunities abroad, regardless of gender. Moreover, statistical analysis shows that female students pursuing overseas postgraduate opportunities in the natural sciences are relatively lacking compared with their male counterparts. This finding might not be limited to the Chinese context, as studies elsewhere suggest a lack of female participation in the STEM fields (Bowen and Rudenstine 2014).

Conclusion
This study has a number of conclusions, some of which perhaps have implications for the empirical research of Bourdieu's convertibility of different forms of capital. Some of these implications are particularly relevant to important issues regarding elite formations in the Chinese context. The main conclusions drawn from the findings are as follows. Firstly, Bourdieu's theoretical standpoint on the convertibility between different forms of capital is particularly relevant in explaining the formation of elite opportunities and prestige through education in China. The three privileged social groups in China-leading cadres and top managers from large enterprises and professionalshave a symbolic association with political, economic and cultural capitals, respectively, although there might be overlapping forms of capital amongst these groups. The children from these social backgrounds have all gained disproportional opportunities in one of the most prestigious universities in China. This finding not only confirms cultural reproduction through elite educational qualifications but also suggests an underlying pattern of the conversion from political and economic capitals to cultural capital to further strengthen their elite status.
However, when the prestige game was extended to the international higher education system, the leading cadres and managerial families had the largest gain, whereas the agricultural families declined most dramatically. Particularly, economic capital is the most consistent predictor of gaining overseas postgraduate offers when all other socioeconomic, meritocratic and demographic factors were considered. This finding suggests the conversion from economic capital by social elites to new forms of cultural capital, which seems to be a new hard currency. This new hard currency combines elite certificates at home and abroad. The formation of this hard currency is significant in several ways. It is not only dressed up in meritocratic measures or elite educational qualifications but is also out of reach for those who have adequate cultural and political capitals but insufficient economic capital. This hard currency' allows elite social groups with strong economic resources to author a new prestige system, which locks out even elite graduates from PKU without adequate economic capital.
Furthermore, the formation of the new hard currency reveals the inherent inconsistencies between the state's meritocracy discourse and its ambition of attracting global talents. Previous studies have highlighted that the meritocracy discourse is integral to China's reform and opening up and further integration into globalisation (Liu 2016). The massive expansion of higher education opportunities and the priority of elite universities are all 'fit' for the purposes of the meritocracy game. On the one hand, the meritocracy discourse induces rural and working-class families to believe that they can improve their social status via the upward social mobility offered by higher education expansion and elite opportunities. On the other hand, the rules of the game are equally applied to those underprivileged groups that failed the exam competitions.
The state's recent ambition of building up double first-class universities and attracting global talents shifted the focus of meritocratic selection based on the Gaokao to prestigious international ranking and status. The state orchestrated the flagship programmes through universities, which is in sync with its ambition to become the world leader in research, science and technology. However, more importantly, it extends the prestige hierarchy to international higher education systems. At the macrolevel, Chinese universities aspire to become leading world universities by succumbing to the international competition fetish. At the individual level, students have also increasingly internalised global prestige and choose the destinations for studying abroad based on the universities' position in the world university ranking. However, this shift of the prestige hierarchy has opened up new space for aspirations and opportunities, which exclude the majority of society, that is, those from working class and agricultural families.
The meritocracy discourse is crucial for social and political stability by successfully inducing disadvantaged families to accept the rules of the game. Thus, the risk of threatening the political order is diminished, and the intensified contradictions amongst different interest groups that emerged during the market reform are mitigated (Liu 2016). However, even with the state's recent ambitious of global talents, it also increases the advantages of overseas universities' qualifications in the domestic labour market. The working class and agricultural families were largely excluded because they have less access to the international universities compared with political, financial and cultural elites. The unintended consequences of the state's global talents discourse might further alienate disadvantaged social groups and deepen the social divide, making the meritocratic rationale less convincing and inclusive.
This study has a number of limitations. The study only included the data from PKU; however, the proportion of elite graduates who chose to study abroad has exceeded 20% of the graduating cohorts in many Project 985 universities. Therefore, the data could not provide a general picture of the patterns of elite mobility from home to abroad. The exact destination patterns in overseas institutions were also impossible to track due to the uncertainties of offers received at the time of the surveys. These limitations suggest further research avenues. For instance, scholars could widen the surveys amongst other elite graduates. Surveys could also include questions on detailed overseas destinations, such as types of universities, fields of study and lengths of the programmes. In undertaking such systematic analysis, scholars could further our understanding of elite mobility from China to abroad through higher education.

Disclosure statement
No potential conflict of interest was reported by the author(s).