To cite this article: Ye Liu (2013) Meritocracy and the Gaokao: a survey study of higher education selection and socio-economic participation in East China, British Journal of Sociology of Education, 34:5-6, 868-887, DOI: 10.1080/01425692.2013.816237
引用本文：Ye Liu (2013) Meritocracy and the Gaokao: a survey study of higher education selection and socioeconomic participation in East China, British Journal of Sociology of Education, 34:5-6, 868-887, DOI: 10.1080/01425692.2013.816237

To link to this article: https://doi.org/10.1080/01425692.2013.816237
本文链接： https://doi.org/10.1080/01425692.2013.816237

Published online: 12 Sep 2013.
在线发表于：2013 年 9 月 12 日。

Submit your article to this journal $\mathit{Z}$$\mathit{Z}$Z\boldsymbol{Z}
向本刊投稿 $\mathit{Z}$$\mathit{Z}$Z\boldsymbol{Z}

Article views: 10374  文章观点10374

View related articles  查看相关文章

Citing articles: 21 View citing articles
引用文章：21 查看引用文章

Keywords: China; meritocracy; the Gaokao; higher education; social stratification
关键词中国；任人唯贤；高考；高等教育；社会分层

Meritocracy has been an enduring theme in modern sociological research. The rise of meritocracy theory was related to dramatic changes in economic, occupational and technological domains after the Second World War in western post-industrial societies (Young 1958; Bell 1973; Goldthorpe 1996). However, there is little empirical evidence to support the notion of a significant trend towards meritocracy in these societies (Breen and Goldthorpe 2001; Whelan and Layte 2002). Intergenerational inheritance has been persistently identified in educational and occupational attainment (Jackson 2007; Arrow, Bowles, and Durlauf 2000). Therefore, meritocracy exists more in belief than reality. The ideology of meritocracy, however, has such an evident attraction that it has been embodied in a variety of political ideologies. It is widely accepted as being an important element of various European centre-left political discourses, particularly of ‘New Labour’ in
任人唯贤是现代社会学研究的一个永恒主题。任人唯贤理论的兴起与第二次世界大战后西方后工业社会在经济、职业和技术领域发生的巨大变化有关（Young，1958 年；Bell，1973 年；Goldthorpe，1996 年）。然而，几乎没有经验证据支持这些社会中出现了明显的任人唯贤的趋势（Breen 和 Goldthorpe，2001 年；Whelan 和 Layte，2002 年）。在教育和职业成就方面，代际继承一直存在（Jackson，2007 年；Arrow、Bowles 和 Durlauf，2000 年）。因此，任人唯贤更多地存在于信念中，而非现实中。然而，任人唯贤的意识形态具有如此明显的吸引力，以至于它已体现在各种政治意识形态中。它被广泛认为是欧洲各种中左翼政治论述的一个重要元素，尤其是欧洲的 "新工党"。

Britain (Goldthorpe and Jackson 2007, 4). The European Union has identified meritocracy as being a key criterion in assessing candidate states for membership of the Union (Vachudova 2005). It has also ostensibly been a key principle in recruitment to jobs in major international organizations such as the IMF and the OECD (Birdsall 2011). Meritocracy exists as a powerful ideology because it represents a type of social selection that, at least nominally, transcends ascriptive boundaries, such as class, gender, ethnicity and nationality, and where there is a positive association between merit and commonly-desired opportunities and rewards.
英国（Goldthorpe 和 Jackson，2007 年，4）。欧盟已将任人唯贤作为评估欧盟候选国的一项关键标准（Vachudova，2005 年）。在国际货币基金组织（IMF）和经合组织（OECD）等主要国际组织的招聘中，任人唯贤显然也是一项关键原则（Birdsall，2011 年）。任人唯贤是一种强大的意识形态，因为它代表了一种社会选择类型，至少在名义上超越了阶级、性别、种族和国籍等描述性界限，并且在优点与共同期望的机会和回报之间存在积极的联系。

Meritocracy, as an ideology with powerful political attraction, was also used by the Chinese Communist Party during China’s transition to the market society. Dramatic changes that took place in 1978 in China have significance beyond the obvious implications of the Reform and Opening-up. More than 30 years of economic growth, with an annual average rate of 9.6% (Zakaria 2008), was accompanied by enduring social problems, including sharply increasing inequality, environmental pollution, and rudimentary health services. The Chinese Communist Party, however, remained relatively secure in the regime. It has been argued that the Party held society together with a transformed ideology, which involved manipulating feelings of patriotism and rejuvenating traditional values and cultures (Spence 2008; Lall and Vickers 2009). Moreover, the promotion of an education-based meritocracy was an important chapter in this transformed ideology.
任人唯贤作为一种具有强大政治吸引力的意识形态，在中国向市场社会转型的过程中也被中国共产党所利用。1978 年，中国发生了翻天覆地的变化，其意义远远超出了改革开放的明显影响。30 多年的经济增长（年均增长率为 9.6%）（Zakaria，2008 年）伴随着持久的社会问题，包括急剧加剧的不平等、环境污染和初级卫生服务。然而，中国共产党在政权中仍然相对稳固。有观点认为，中国共产党以一种转型的意识形态来维系社会，这种意识形态包括操纵爱国主义情感、复兴传统价值观和文化（Spence，2008 年；Lall 和 Vickers，2009 年）。此外，促进以教育为基础的任人唯贤也是这一转型意识形态的重要组成部分。

This article examines how meritocracy was used by the Chinese Communist Party to deal with dramatic socio-economic and political circumstances during the reform period, and how higher education selection became an ideal vehicle with which the Party could associate itself with the ideology of meritocracy. I begin by outlining the research area, which investigates the relationship between education and meritocracy, before moving on to different theoretical perspectives applied in the Chinese context. I then address some attributes of China as a case for investigating the implications of the ideology of meritocracy. I use a survey study conducted in Anhui and Zhejiang (East China) in 2007 to examine the extent to which contemporary higher education selection was based on meritocratic criteria. This smallscale survey was used as the primary data source mainly because there is a lack of national data on students’ socio-economic backgrounds and other demographic information and because it is difficult to collect at a comprehensive level the data that represent all Chinese provinces. This survey study is, of course, inevitably limited as a means of mapping out a general picture of social inequality in higher education in contemporary China. However, the article is able to provide a suggestive analysis with regard to socio-economic participation in higher education in East China. Methodological issues concerning this survey study will be specifically addressed in the following sections. A statistical analysis is then presented to argue the
本文探讨了中国共产党在改革时期如何利用任人唯贤来应对剧烈的社会经济和政治环境，以及高等教育选拔如何成为一种理想的载体，使中国共产党能够将自己与任人唯贤的意识形态联系起来。我首先概述了研究领域，即调查教育与任人唯贤之间的关系，然后介绍了在中国背景下应用的不同理论视角。然后，我将中国的一些特点作为研究任人唯贤思想影响的案例。我利用 2007 年在安徽和浙江（华东）进行的一项调查研究，来考察当代高等教育选拔在多大程度上是基于择优标准的。之所以采用这项小规模调查作为主要数据来源，主要是因为缺乏有关学生社会经济背景和其他人口统计信息的全国性数据，而且很难全面收集代表中国所有省份的数据。当然，这项调查研究在描绘当代中国高等教育中社会不平等的总体图景方面难免有其局限性。然而，本文能够对华东地区高等教育中的社会经济参与情况进行提示性分析。下文将具体讨论本调查研究的方法问题。然后，本文将通过统计分析来论证
implications of the meritocracy ideology on the Party’s overall development strategy and on the life-chances of different social groups.
任人唯贤思想对党的整体发展战略和不同社会群体生活机会的影响。

Meritocracy has been a controversial sociological term. Broadly speaking, meritocracy means ‘a large-scale social system in which a positive relationship exists between merit and such commonly desired values as income, power, and prestige’ (Krauze and Slomczynski 1985, 623). Merit has been interpreted as educational qualifications (Bell 1973) or intelligence plus effort (Young 1958, 94). Despite the disputed definition of merit, formal education is used as the key variable to measure meritocracy and the educational system is regarded as a selection mechanism that measures and rewards merit (Olneck and Crouse 1979; Goldthorpe 1996). This article focuses primarily on education-based meritocracy instead of a multifaceted social selection, mainly because of the central role of education in examining the process of social status attainment from social origin to destination in the meritocracy thesis (Lipset and Zetterberg 1959; Bell 1973).
任人唯贤一直是一个有争议的社会学术语。从广义上讲，"功绩制 "指的是 "一种大规模的社会制度，在这种制度中，功绩与收入、权力和声望等人们普遍期望的价值之间存在着一种积极的关系"（Krauze and Slomczynski 1985, 623）。绩优被解释为学历（Bell，1973 年）或智力加努力（Young，1958 年，94 年）。尽管对 "优点 "的定义存在争议，但正规教育被用作衡量 "优点 "的关键变量，教育系统被视为衡量和奖励 "优点 "的选拔机制（Olneck 和 Crouse，1979 年；Goldthorpe，1996 年）。本文主要关注以教育为基础的择优，而不是多方面的社会选择，这主要是因为在择优论中，教育在研究从社会起源到目的地的社会地位获得过程中发挥着核心作用（Lipset 和 Zetterberg，1959 年；Bell，1973 年）。

Educational meritocracy is often used as a notion with two complementary aspects: how social origin impacts on educational opportunities and achievement; and how merit measured by educational outcomes is related to social destination, which specifically refers to students’ destination in types of higher education in this article. As far as meritocratic selection is concerned, educational attainment should not only reflect an individual’s ability (or intelligence plus effort) rather than his or her social origin, but it also should promise opportunities and rewards. According to meritocracy theory, an education-based meritocracy can be said to exist where educational attainment is unrelated to social origin and a decisive influence on social destination.
教育择优经常被用作一个概念，它包含两个相辅相成的方面：社会出身如何影响教育机会和成就；以教育成果衡量的择优与社会去向（本文中特指学生接受高等教育的去向）如何相关。就择优选拔而言，教育成就不仅应反映个人的能力（或智力加努力），而不是其社会出身，而且还应承诺机会和回报。根据择优选拔理论，如果教育程度与社会出身无关，并对社会去向产生决定性影响，就可以说存在以教育为基础的择优选拔。

Regarding the first aspect of educational meritocracy - that is, how social origin impacts on education - socio-economic characteristics are treated by Breen and Jonsson (2005) and Jackson (2007) as being the crucial indicator for assessing barriers to meritocratic selection in education. However, these studies mainly focus on western industrial or post-industrial societies. Extending this theoretical perspective to the Chinese context requires highlighting some context-related issues. Most scholarship on educational selection in the context of China has followed the tradition of development theory, which has predominantly focused on gender and regional disparities as key factors that generated glaring inequality in educational attainment (Li and Tsang 2003; Hannum and Wang 2006). Empirical evidence has tested and further confirmed the standpoints of development theory, arguing that the uneven distribution of educational resources and the socially/culturally differentiated expectations of males and females are factors in educational selection based on meritocratic principles (Li and Tsang 2003; Hannum and
关于教育择优的第一个方面，即社会出身如何影响教育，Breen 和 Jonsson（2005 年）以及 Jackson（2007 年）将社会经济特征视为评估教育择优障碍的关键指标。然而，这些研究主要侧重于西方工业社会或后工业社会。将这一理论视角推广到中国的背景下，需要强调一些与背景相关的问题。大多数关于中国教育选拔的学术研究都沿袭了发展理论的传统，主要关注性别和地区差异，将其视为造成教育程度严重不平等的关键因素（Li 和 Tsang，2003 年；Hannum 和 Wang，2006 年）。经验证据检验并进一步证实了发展理论的观点，认为教育资源的不均衡分配以及社会/文化对男性和女性的不同期望是基于择优原则的教育选择的因素（李和曾，2003 年；汉纳姆和王，2006 年）。

Park 2007; Hannum, Wang, and Adams 2008). Little research has been conducted to examine specifically the extent to which socio-economic status impacts on educational selection in contemporary China. Hence, in this article variables such as geographical origin and gender will be used as characteristics of an individual’s demographic background along with socioeconomic status when evaluating educational selection in China.
Park 2007；Hannum、Wang 和 Adams 2008）。关于社会经济地位对当代中国教育选择的影响程度，鲜有专门研究。因此，在本文中，在评估中国的教育选择时，地域来源和性别等变量将与社会经济地位一起作为个人人口背景的特征。

Theories on the in/equality of educational opportunities illustrate further the implication of social origin for education. Maximally Maintained Inequality (MMI) theory provides one perspective on the implications of the expansion of educational opportunities for meritocratic selection in education. According to this theory, the effect of social class on educational attainment should decline after a given level of education becomes universal for the upper social class (Raftery and Hout 1993). While they have reached a threshold, the expansion of educational opportunities allows lower social groups to advance. MMI theory argues that socio-economic characteristics impact more significantly on educational attainment at the higher level rather than at the basic level, because this is where the ceiling is reached. However, MMI theory was developed in the context of western industrial societies and may not apply in the same way in the context of China. Universal access to compulsory education in China was provided in urban areas prior to the rural areas. Data on access to compulsory education by the age cohort ( $7-16$$7-16$7-167-16 year olds) from 1993 to 2004 show a 10 -year lag between the rural and urban areas in terms of enrolment rates; that is, $96\mathrm{\%}$$96\mathrm{\%}$96%96 \% of the rural age cohort in 2004 compared with $95\mathrm{\%}$$95\mathrm{\%}$95%95 \% of the urban cohort in 1993 (CHNS 2004). This urban-rural lag in the provision of universal access to compulsory education complicates the picture of the socio-economic differences in access to educational opportunities.
关于教育机会不平等的理论进一步说明了社会出身对教育的影响。最大限度地维持不平等（MMI）理论从一个角度说明了扩大教育机会对教育中择优选拔的影响。根据这一理论，在上层社会普及了一定程度的教育之后，社会阶层对教育程度的影响就会下降（Raftery 和 Hout，1993 年）。当他们达到一个临界点时，教育机会的扩大使较低的社会群体得以提升。多指标类集理论认为，社会经济特征对较高层次教育程度的影响比对基础教育程度的影响更大，因为这正是达到天花板的地方。然而，MMI 理论是在西方工业社会背景下提出的，在中国可能并不适用。在中国，城市地区先于农村地区普及义务教育。1993 年至 2004 年按年龄组（ $7-16$$7-16$7-167-16 岁）划分的义务教育入学率数据显示，农村和城市的入学率相差 10 年，即 2004 年农村年龄组的入学率为 $96\mathrm{\%}$$96\mathrm{\%}$96%96 \% ，而 1993 年城市年龄组的入学率为 $95\mathrm{\%}$$95\mathrm{\%}$95%95 \% （中国国家统计局，2004 年）。在普及义务教育方面存在的这种城乡差别，使获得教育机会方面的社会经济差异情况更加复杂。

The lag has significant implications for the patterns of access to postcompulsory education, particularly higher education opportunities. It has been argued that universal access to compulsory education up to secondary schooling provided a crucial demographic base for the expansion of higher education in the experiences of most post-War western societies (Shavit and Blossfeld 1993). In China, higher education has expanded at an unprecedented rate since the mid-1990s, a time prior to rural universal
这种滞后对义务教育后的教育模式，特别是高等教育机会的获得具有重大影响。有人认为，在大多数战后西方社会的经验中，普及义务教育直至中学教育为高等教育的扩展提供了重要的人口基础（Shavit 和 Blossfeld，1993 年）。在中国，自 20 世纪 90 年代中期以来，高等教育以前所未有的速度发展。

Table 1. Age participation ratio of students in higher education in China from 1990 to 2006.
表 1.1990 至 2006 年中国高等教育学生的年龄参与率。

Source: China Statistics Year Book 1991, 1999, 2007 (National Bureau of Statistics 1991, 1999, 2007).
《资料来源中国统计年鉴 1991、1999、2007》（国家统计局，1991、1999、2007）。
access to compulsory education. Table 1 shows the expansion of student numbers as a percentage of the 20-24 age cohorts from 1990 to 2006. The number of students in higher education increased from 1.58 million in 1990 to over 23 million in 2006, rising from 1.5 to $21\mathrm{\%}$$21\mathrm{\%}$21%21 \% of the $20-24$$20-24$20-2420-24 age cohort. The Chinese pattern of the expansion of educational opportunities, with the rural-urban lag in providing universal access to compulsory education and a massive expansion of higher education, provides an interesting case to examine the arguments from MMI theory.
接受义务教育的机会。表 1 显示了 1990 至 2006 年间学生人数的增长情况，占 20-24 岁年龄组的百分比。接受高等教育的学生人数从 1990 年的 158 万人增加到 2006 年的 2300 多万人，占 $20-24$$20-24$20-2420-24 年龄组的比例从 1.5%上升到 $21\mathrm{\%}$$21\mathrm{\%}$21%21 \% 。中国在普及义务教育方面的城乡滞后和高等教育的大规模扩张为研究多指标类集理论的论点提供了一个有趣的案例。

The second aspect of educational meritocracy - that is, how merit is related to students’ destination - is thoroughly elaborated in Effectively Maintained Inequality (EMI) theory. EMI theory, contrary to MMI theory, argues that the effects of social origin on educational attainment would not decrease, even if a given level of education became universal for advantaged social classes. Instead, more advantaged social classes would seek to secure qualitatively better types of education in the face of the expansion of educational opportunities for other groups. A large number of studies have tested the EMI standpoint with empirical evidence (Lucas 2001; Zimdars 2007). A correspondence between students’ socio-economic characteristics, such as socio-economic status and parental education, and their destinations, in types of universities, is suggested from research in a variety of social contexts (Ayalon and Yogev 2005; Duru-Bellat, Kieffer, and Reimer 2008).
有效维持不平等（EMI）理论深入阐述了教育择优的第二个方面，即择优与学生去向的关系。EMI 理论与 MMI 理论相反，它认为，即使特定的教育水平在优势社会阶层中得到普及，社会出身对教育程度的影响也不会减少。相反，面对其他群体受教育机会的扩大，更有优势的社会阶层会寻求质量更好的教育类型。大量研究用实证检验了 EMI 观点（Lucas，2001 年；Zimdars，2007 年）。在各种社会背景下进行的研究表明，学生的社会经济特征（如社会经济地位和父母教育程度）与他们的目的地（大学类型）之间存在对应关系（Ayalon 和 Yogev，2005 年；Duru-Bellat、Kieffer 和 Reimer，2008 年）。

Elaborating this theoretical standpoint in the context of China requires a specific examination of the pattern of expansion of higher education. The expansion of higher education opportunities has been accompanied by an increasingly stratified system. The project of building ‘world-class’ ${}^1$${}^1$^(1){ }^{1} universities prioritized the provision of funding and resources to elite universities (Tier One or the 985 universities) and key universities (Tier Two or the 211 universities). Third-tier and fourth-tier universities, ${}^2$${}^2$^(2){ }^{2} particularly vocational and technological institutions at the provincial level, have been poorly staffed and under-funded. They have had to struggle for resources from the government, and consequently are highly subjected to the vagaries of the unfettered market. This mixed pattern of expansion of higher education in China suggests a qualitative difference in addition to the quantitative difference in the expanded higher education system. Hence, this provides an interesting case to test the EMI argument of the qualitative difference by examining the role of education in mediating intergenerational inheritance and the expansion of educational opportunities in realizing a meritocratic selection.
要在中国的背景下阐述这一理论观点，需要对高等教育的扩张模式进行具体研究。在扩大高等教育机会的同时，高等教育体系也日益分层。建设 "世界一流" ${}^1$${}^1$^(1){ }^{1} 大学的项目优先向精英大学（一级或 985 大学）和重点大学（二级或 211 大学）提供资金和资源。三级和四级大学， ${}^2$${}^2$^(2){ }^{2} 特别是省级职业技术院校，人员配备不足，资金不足。它们不得不努力从政府那里争取资源，因此极易受到不受约束的市场的影响。中国高等教育扩张的这种混合模式表明，高等教育系统的扩张除了数量上的差异外，还存在质量上的差异。因此，通过研究教育在代际传承中的中介作用以及教育机会的扩大在实现择优选拔中的作用，这为检验质的差异的 EMI 论点提供了一个有趣的案例。

China offers some attractive attributes as a case through which to examine the links between the expansion of educational opportunities, the ideology of meritocracy and social inequality. Firstly, China’s transition from a
中国提供了一些极具吸引力的特质，可作为研究扩大教育机会、任人唯贤的意识形态和社会不平等之间联系的案例。首先，中国从

Socialist society to a ‘Socialist market economy with Chinese characteristics’ since 1978 has had direct effects on the means of social selection. Political affiliation and loyalty were characteristics of social selection in a socialist society to maintain the Communist party as the ruling class (Unger 1982, 3). Recruitment to higher education institutions used to give preference to those from the ‘red’ classes, including workers, peasants and party members (Li and Walder 2001, 1387). The ‘Reform and Opening-up’ in 1978 not only marked the economic collapse of Communist production in China, but also the decline of the Party’s predominantly political affiliationbased social selection. The Gaokao was restored as a key mechanism of structuring higher education opportunities, and the academic performance in the Gaokao, instead of political affiliation, became the decisive factor in access to higher education. This transition in China offers an unusual opportunity to investigate the extent to which the Gaokao selection is based on meritocratic criteria.
自 1978 年以来，从社会主义社会到 "有中国特色的社会主义市场经济 "对社会选择的方式产生了直接影响。在社会主义社会中，政治归属和忠诚是社会选拔的特征，以维护共产党作为统治阶级的地位（Unger，1982 年，3）。高等院校在招生时优先考虑来自 "红色 "阶层的学生，包括工人、农民和党员（Li and Walder 2001, 1387）。1978 年的 "改革开放 "不仅标志着共产党在中国的生产在经济上的崩溃，也标志着党以政治派别为主导的社会选拔方式的衰落。高考作为高等教育机会结构的关键机制得以恢复，高考成绩而非政治背景成为接受高等教育的决定性因素。中国的这一转变为研究高考选拔在多大程度上基于择优标准提供了一个不同寻常的机会。

Secondly, coinciding with the dramatic economic changes was the emergence of a more stratified social structure. The pre-Reform society was characterized by egalitarian social groups, which included workers, peasants, soldiers and the Party cadres. However, the opportunity structure that had been dominated by the Communist cadres’ hierarchy was opened up to a much wider group of institutions and individuals (Nee and Matthews 1996). Those individuals who had seized the new opportunities that emerged from the redistribution of capital and resources during the market transition formed new social groups, such as entrepreneurs, professionals and cadres transformed to businessmen (Huang 1993; Qin 1999; Duckett 2001). The ‘homogeneous peasants class’ (Parish 1975) became stratified into different types of agricultural workers in the rural areas and, in a more significant way, between these and peasant migrants to urban areas (Harvey 2005). Moreover, the working class was differentiated in many ways, including between wage labour in the private sector, layoff labour, and unprotected labour in the state-owned enterprises (Walker and Buck 2007). The emergence of a stratified social structure during the transition makes China an interesting case for examining the extent to which social origin impacts on opportunity structures.
其次，与经济巨变同时出现的是更加分层的社会结构。改革前社会的特点是平等社会群体，包括工人、农民、军人和党的干部。然而，由共产党干部等级制度主导的机会结构向更广泛的机构和个人群体开放（Nee 和 Matthews，1996 年）。那些抓住了市场转型期资本和资源再分配带来的新机遇的个人形成了新的社会群体，如企业家、专业人士和转型为商人的干部（Huang，1993 年；Qin，1999 年；Duckett，2001 年）。同质的农民阶级"（Parish，1975 年）在农村地区被分化为不同类型的农业工人，更重要的是，在他们与向城市地区迁移的农民之间也出现了分层（Harvey，2005 年）。此外，工人阶级在许多方面也被区分开来，包括私营部门的雇佣劳动、下岗劳动和国有企业中不受保护的劳动（Walker 和 Buck，2007 年）。转型期出现的分层社会结构使中国成为研究社会出身对机会结构影响程度的一个有趣案例。

This article examines the extent to which higher education selection in contemporary China is based on meritocratic criteria. A hypothesis concerning two aspects of education-based meritocracy developed from the previous analysis is that a meritocratic selection would suggest a weak association between a student’s social origin and his or her academic performance, and that the academic performance would lead to corresponding destinations in higher education regardless of social origin. First, a series of hypotheses can
本文探讨了当代中国高等教育选拔在多大程度上是基于择优标准的。在前文的分析基础上，我们提出了关于基于教育的择优标准的两个方面的假设，即择优选拔将表明学生的社会出身与其学业成绩之间的关联较弱，而且无论社会出身如何，学业成绩都将导致相应的高等教育去向。首先，我们可以提出一系列假设
be formulated regarding the influence of socio-economic background and socio-demographic characteristics on the Gaokao performance. Moreover, socio-economic and demographic backgrounds will be examined along with the Gaokao performance in relation to students’ destination in types of universities.
社会经济背景和社会人口特征对高考成绩的影响。此外，社会经济背景和人口统计背景还将与学生的高考成绩和大学类型相关联。

Because no single set of national data is available to allow a comprehensive examination of undergraduates’ socio-economic characteristics, I conducted a survey study involving around 960 first-year undergraduate students in eight different types of universities in two provinces (Anhui and Zhejiang) in 2007. ${}^3$${}^3$^(3){ }^{3} First-hand empirical research is always subject to some methodological limitations. A possible criticism of this study is that all participants have successfully survived in the Gaokao competition. It is therefore possible that this analysis is unable to provide a full picture of the survival rates of different social groups in the higher education selection. Hence, prior to investigating the socio-economic characteristics of the surveyed students, it is necessary to provide a general picture of the survival rate of this birth cohort (1986-1989) through various selection points in the education system. This statistical analysis is generated from data provided in the China Statistical Year Book 2006. Among the sampled students, around $90\mathrm{\%}$$90\mathrm{\%}$90%90 \% were born between 1986 and 1989. At the national level, higher education students are those who successfully survived the junior secondary selection and the senior secondary selection. These survivors account for only around $20\mathrm{\%}$$20\mathrm{\%}$20%20 \% of the correspondent age cohort. Moreover, only $11.6\mathrm{\%}$$11.6\mathrm{\%}$11.6%11.6 \% of these survivals secure a Bachelor degree in higher education institutions (see Table 2). Among these four-year programme higher education institutions, the 985 or Tier One universities are highly selective, recruiting only $3.7\mathrm{\%}$$3.7\mathrm{\%}$3.7%3.7 \% of students who survived the competition in the Gaokao for four-year programme opportunities. The 211 or Tier Two universities recruited around $11\mathrm{\%}$$11\mathrm{\%}$11%11 \% of such students. Institutions at Tier Three and Tier Four enrolled the
由于没有一套全国性的数据可以全面考察本科生的社会经济特征，笔者于2007年对两省（安徽和浙江）八所不同类型高校的约960名本科一年级学生进行了调查研究。 ${}^3$${}^3$^(3){ }^{3} 第一手的实证研究总是会受到一些方法论上的限制。本研究可能受到的一个批评是，所有参与者都在高考竞争中成功存活下来。因此，本分析有可能无法全面反映不同社会群体在高等教育选拔中的存活率。因此，在调查被调查学生的社会经济特征之前，有必要对这批出生的学生（1986-1989 年）在教育系统的各个选拔点的存活率进行总体描述。本统计分析来自《中国统计年鉴 2006》提供的数据。在抽样调查的学生中，约 $90\mathrm{\%}$$90\mathrm{\%}$90%90 \% 出生于 1986 年至 1989 年之间。在全国范围内，高等教育学生是指成功通过初中选拔和高中选拔的学生。这些幸存者只占相应年龄组的 $20\mathrm{\%}$$20\mathrm{\%}$20%20 \% 左右。此外，这些幸存者中只有 $11.6\mathrm{\%}$$11.6\mathrm{\%}$11.6%11.6 \% 能在高等教育机构获得学士学位（见表 2）。在这些四年制高等院校中，985 所或一级大学的选择性很强，只招收 $3.7\mathrm{\%}$$3.7\mathrm{\%}$3.7%3.7 \% 名在高考竞争中幸存下来的学生，以获得四年制课程的机会。211所二本大学招收了约 $11\mathrm{\%}$$11\mathrm{\%}$11%11 \% 名此类学生。 第三级和第四级院校招收了

Table 2. Education careers of those born between 1986 and 1988 in China.
表 2.中国 1986 年至 1988 年出生人口的教育经历。

Table 3. Selection rates for different types of higher education institutions with four-year Bachelor programmes in 2006.
表 3.2006 年不同类型高等教育机构四年制学士学位课程的入选率。

Source: National Bureau of Statistics (2007).
资料来源：国家统计局（2007 年）：资料来源：国家统计局（2007 年）。
Note: ${}^{\text{a}}$${}^{\text{a }}$^("a "){ }^{\text {a }} The selection rate was calculated by a number of indicators. The number of new entrants for the 985 universities was added together from the new entry report published by each individual 985 institution. This number was then divided by the number of Bachelor’s degree programme entrants for 2006 given in China Statistical Year Book 2007 (National Bureau of Statistics 2007).
注： ${}^{\text{a}}$${}^{\text{a }}$^("a "){ }^{\text {a }} 选拔率由多项指标计算得出。根据各 985 院校公布的新生入学报告，将 985 院校的新生入学人数相加。然后将这一数字除以《2007 年中国统计年鉴》（国家统计局，2007 年）中提供的 2006 年学士学位课程入学人数。
majority of students, accounting for around $85\mathrm{\%}$$85\mathrm{\%}$85%85 \% of students who survived the Gaokao in 2006 (see Table 3).
占大多数，约 $85\mathrm{\%}$$85\mathrm{\%}$85%85 \% 占 2006 年高考幸存学生的 $85\mathrm{\%}$$85\mathrm{\%}$85%85 \% （见表 3）。

The questionnaires were administered in eight higher education institutions in Anhui and Zhejiang. Four universities were chosen in each province. These higher education institutions were selected from each tier of the higher education system. Table 4 demonstrates how these eight universities were distributed to cover all tiers of the higher education system across two provinces. The sampling strategy was designed to correspond to the selection rates of different types of universities shown in Table 3. Over-sampling or under-sampling in a survey study will lead to a distortion of representation of surveyed population and subsequently affects the quality of the data collection. Around 40 students were sampled in the two 985 universities and 400 students were sampled in Tier Three and Tier Four institutions respectively, a ratio of sampling between two types of universities corresponding to the selection rates of the 985 universities and Tier Three and Tier Four universities after the 2006 Gaokao. The original number of the participants in this study was 960 , with 480 students in Anhui and 480 in Zhejiang. A total 858 of the questionnaires were used in the final analysis after an initial test for credibility of the answers given in each questionnaire. ${}^4$${}^4$^(4){ }^{4} Table 5 shows the original figures of the participants in each university with the final sample figures.
问卷调查在安徽和浙江的八所高校进行。每个省选择了四所高校。这些高校分别选自高等教育系统的各个层级。表 4 显示了这八所高校的分布情况，以涵盖两省高等教育系统的所有层级。抽样策略的设计符合表 3 所示的不同类型大学的选取率。在调查研究中，抽样过多或过少都会导致被调查人群的代表性失真，进而影响数据收集的质量。在两所 985 大学中分别抽取了约 40 名学生，在三四线院校中分别抽取了约 400 名学生，两类大学的抽样比例与 2006 年高考后 985 大学和三四线大学的入选率相对应。本研究的原始参与者人数为 960 人，其中安徽学生 480 人，浙江学生 480 人。在对每份问卷的答案进行可信度初步测试后，共有 858 份问卷被用于最终分析。 ${}^4$${}^4$^(4){ }^{4} 表 5 显示了各高校参与者的原始数据和最终样本数据。

Table 4. Sampling strategy for types of higher education institutions.
表 4.高等教育机构类型的抽样策略。

Table 5. Sample strategy for numbers of students in different types of universities: 858 out of 960 .
表 5.不同类型大学学生人数的抽样策略：960 人中有 858 人。

The obvious limitation of this survey is the response rate and the representativeness of the surveyed students. ${}^5$${}^5$^(5){ }^{5} The main focus of this study is to examine how socio-economic status affected students’ academic performance and how students’ academic attainment determines their destinations in different types of university. Hence, the survey study was conducted for the specific purpose of collecting students’ information on socio-economic backgrounds. In addition to the survey data, another data source was employed to provide a comparison of the general patterns of the socio-economic composition at the national level. The data on socio-economic status came from Lu Xueyi’s statistical study of contemporary social stratification in China (Lu 2010). Details of the comparison are given in this section when they are introduced.
本调查的明显局限性在于受访学生的回复率和代表性。 ${}^5$${}^5$^(5){ }^{5} 本研究的主要重点是探讨社会经济地位如何影响学生的学业成绩，以及学生的学业成绩如何决定他们在不同类型大学的去向。因此，调查研究的具体目的是收集学生的社会经济背景信息。除调查数据外，还采用了另一种数据来源，以比较全国社会经济构成的一般模式。有关社会经济地位的数据来自陆学艺的《当代中国社会分层统计研究》（陆学艺，2010 年）。比较的详细情况将在本节介绍时给出。

The first independent variable used is socio-economic status. Socio-economic status is certainly the most frequently used term in sociological research, but it is conceptualized differently in different national contexts. The survey used the classification into 10 socio-economic groups developed by Lu Xueyi in his report on contemporary social class in China (Table 6) (Lu 2010). Socio-economic status was re-coded into five categories in the following statistical analysis: leading cadres and managers; private entrepreneurs; professional and clerical employees; working class; and agricultural and unemployed/underemployed workers. Parental educational level is another commonly used indicator of socio-economic characteristics in sociological research (OECD 2007; Duru-Bellat, Kieffer, and Reimer 2008). ${}^6$${}^6$^(6){ }^{6} The details of coding parental educational levels are included in the notes.
使用的第一个自变量是社会经济地位。社会经济地位无疑是社会学研究中最常用的术语，但在不同的国家背景下，其概念有所不同。本次调查采用了陆学艺在《中国当代社会阶层报告》中提出的 10 个社会经济群体的划分方法（表 6）（陆学艺，2010 年）。在接下来的统计分析中，社会经济地位被重新编码为五类：领导干部和管理人员；私营企业主；专业和文职雇员；工人阶级；农业和失业/半失业工人。父母受教育程度是社会学研究中另一个常用的社会经济特征指标（经合组织，2007 年；Duru-Bellat、Kieffer 和 Reimer，2008 年）。 ${}^6$${}^6$^(6){ }^{6} 父母教育水平的编码细节见注释。

Previous discussions suggest geographical origin as a key indicator of socio-economic characteristics in the Chinese context. This study uses three categories defining students’ geographical origin of birth - cities, counties and the rural areas. Types of schooling are considered another indicator of socio-demographic characteristics. Schooling has been a key variable in understanding socio-economic participation in higher education in sociological research (Zimdars 2007; Schagen, Davies, and Rudd 2002). Chinese senior secondary schooling is categorized into model/key schools and regular secondary schools. ${}^7$${}^7$^(7){ }^{7}
以往的讨论表明，地域来源是衡量中国社会经济特征的一个关键指标。本研究将学生的出生地分为三类--城市、县城和农村。学校教育类型被认为是社会人口特征的另一个指标。在社会学研究中，学校教育一直是理解社会经济参与高等教育的一个关键变量（Zimdars，2007 年；Schagen、Davies 和 Rudd，2002 年）。中国的高中教育分为示范/重点中学和普通中学。 ${}^7$${}^7$^(7){ }^{7}

Table 6. Comparison between the surveyed population and the corresponding composition in the general population.
表 6.被调查人口与普通人口相应构成的比较。

Source: Data on socio-economic participation in higher education are generated from my empirical work; data on the composition of the corresponding population come from Lu (2010). Note: ${}^{\text{a}}$${}^{\text{a }}$^("a "){ }^{\text {a }} The first two socio-economic groups are combined; that is, the leading cadres/governmental officials and executive personnel.
资料来源社会经济参与高等教育的数据来自笔者的实证研究；相应人口构成的数据来自 Lu (2010)。注： ${}^{\text{a}}$${}^{\text{a }}$^("a "){ }^{\text {a }} 前两个社会经济群体是合并的，即领导干部/政府官员和行政人员。

The elaboration of the dependent variables - the Gaokao performance and the destination in types of universities - follows the previous discussions on the two aspects of educational meritocracy. First, the Gaokao scores will be treated as the measure of an individual’s merit in this article. The relation between socio-economic and demographic characteristics and Gaokao scores will illustrate the extent to which higher education selection is based on meritocratic principles. The data on the Gaokao scores came from the survey study; and answers about the Gaokao scores in the questionnaires were tested for their credibility by combining the published information on cutting-off points in each province and the detailed final enrolment reports in each university. ${}^8$${}^8$^(8){ }^{8} The detailed coding of the Gaokao scores is explained in the regression analysis section. Second, students’ destination in types of universities will be further examined in relation to their Gaokao performance as well as their socio-economic and demographic variables. The educational meritocracy hypothesis suggests that students’ destination in types of universities should be determined by their Gaokao scores regardless of their socio-economic backgrounds. The destination in types of universities will mainly be distinguished between elite universities (the 985 and 211 institutions) and non-elite universities (provincial and vocational institutions).
对因变量--高考成绩和大学类型的去向--的阐述沿袭了前文对教育择优的两个方面的讨论。首先，本文将把高考成绩作为衡量个人优劣的标准。社会经济和人口特征与高考分数之间的关系将说明高等教育选拔在多大程度上是基于择优原则的。有关高考分数的数据来自调查研究，问卷中有关高考分数的回答则通过结合各省公布的分数线信息和各高校详细的最终招生报告来检验其可信度。 ${}^8$${}^8$^(8){ }^{8} 高考分数的具体编码方法将在回归分析部分进行说明。其次，我们将结合学生的高考成绩、社会经济变量和人口统计变量，进一步研究学生的大学类型去向。教育择优假说认为，无论学生的社会经济背景如何，他们在各类大学中的去向应由高考成绩决定。大学类型的去向将主要区分为精英大学（"985 "和 "211 "院校）和非精英大学（省属和高职院校）。

This section presents a descriptive statistical analysis of the general pattern of participation in higher education by different socio-economic groups. Table 6 demonstrates the pattern of socio-economic participation of the surveyed population; and compares this pattern with the corresponding composition of the general population. First, children from two socio-economic groups - leading cadres and managerial personnel - represented $1.5\mathrm{\%}$$1.5\mathrm{\%}$1.5%1.5 \% of the participants in comparison with $3.6\mathrm{\%}$$3.6\mathrm{\%}$3.6%3.6 \% of that group in the whole population. Children of private entrepreneurs accounted for around $4\mathrm{\%}$$4\mathrm{\%}$4%4 \% of the surveyed population, compared with $2.6\mathrm{\%}$$2.6\mathrm{\%}$2.6%2.6 \% in the general population. Children of professionals and clerical workers had a large representation, accounting for nearly one-third of the surveyed students in comparison with the $13.3\mathrm{\%}$$13.3\mathrm{\%}$13.3%13.3 \% of that group in the whole population. Children of working-class parents were under-represented, accounting for around $18\mathrm{\%}$$18\mathrm{\%}$18%18 \% of surveyed students by comparison with the $24.8\mathrm{\%}$$24.8\mathrm{\%}$24.8%24.8 \% of that group in the whole population. This selectivity was most favourable to students from professional families while most unfavourable to students from manufacturing working-class backgrounds. Noticeably, students from agricultural families had a relatively large representation in the survey. However, the agricultural families are still under-represented when compared with their proportion in the whole population. The following sections further discuss the socio-economic patterns that influence the attainment of higher education, by detailing Gaokao performance and subsequent destination in higher education.
本节对不同社会经济群体接受高等教育的总体情况进行了描述性统计分析。表 6 显示了被调查人口的社会经济参与模式，并将这一模式与一般人口的相应构成进行了比较。首先，来自两个社会经济群体--领导干部和管理人员--的子女在受访者中所占比例为 $1.5\mathrm{\%}$$1.5\mathrm{\%}$1.5%1.5 \% ，而该群体在总人口中所占比例为 $3.6\mathrm{\%}$$3.6\mathrm{\%}$3.6%3.6 \% 。私营企业主的子女约占被调查人口的 $4\mathrm{\%}$$4\mathrm{\%}$4%4 \% ，而一般人口中这一群体的子女占 $2.6\mathrm{\%}$$2.6\mathrm{\%}$2.6%2.6 \% 。专业人员和文秘人员的子女所占比例较大，占被调查学生的近三分之一，而该群体在总人口中所占比例为 $13.3\mathrm{\%}$$13.3\mathrm{\%}$13.3%13.3 \% 。工人阶级父母的子女所占比例较低，约占被调查学生的 $18\mathrm{\%}$$18\mathrm{\%}$18%18 \% ，而该群体在总人口中所占比例为 $24.8\mathrm{\%}$$24.8\mathrm{\%}$24.8%24.8 \% 。这种选择性对来自专业家庭的学生最有利，而对来自制造业工人阶级背景的学生最不利。值得注意的是，来自农业家庭的学生在调查中所占比例相对较大。然而，与农业家庭在整个人口中所占的比例相比，农业家庭的比例仍然偏低。下文将通过详细介绍高考成绩及其后的升学去向，进一步讨论影响学生接受高等教育的社会经济模式。

Table 7 reports the results from a series of simple logistic regression analyses of the log-odds of independent variables on the Gaokao performance where the best academic outcome was coded as one, and a low performance in the Gaokao coded was as zero. ${}^9$${}^9$^(9){ }^{9}
表 7 报告了一系列简单逻辑回归分析的结果，即自变量与高考成绩的对数之比，其中学习成绩最好的结果编码为 "1"，高考成绩差的结果编码为 "0"。 ${}^9$${}^9$^(9){ }^{9}

The logistic regression predicts the $\log$$\log$log\log odds that an observation will have an indicator equal to one. The odds of being one of the top performers in the Gaokao is defined as the ratio of the probability that a student achieves scores in the highest threshold to the probability that the candidate is on the lower thresholds of the Gaokao performance. The best academic performance is understood as the highest threshold of the Gaokao. The highest threshold in the 2006 Gaokao consists of those who obtained more than 560 points out of a total 750 . Model 1 shows the net effect of socioeconomic status on Gaokao performance. In this model, students from professional backgrounds tended to achieve higher scores than students from agricultural families. Students from leading cadres and business executive backgrounds and from entrepreneur backgrounds do not seem to have significant advantage in the Gaokao over students from agricultural families.
逻辑回归预测的是一个观测值的指标等于 1 的 $\log$$\log$log\log 几率。成为高考状元的几率被定义为学生取得最高门槛分数的概率与考生处于高考成绩较低门槛的概率之比。学习成绩最好可以理解为高考的最高门槛。2006 年高考的最高门槛是在总分 750 分中获得 560 分以上的考生。模型 1 显示了社会经济地位对高考成绩的净影响。在该模型中，专业背景的学生往往比来自农业家庭的学生取得更高的分数。与来自农业家庭的学生相比，来自领导干部和企业高管背景的学生以及来自企业家背景的学生似乎在高考中没有明显优势。

Table 7. Simple logit regression for achieving highest thresholds in the 2006 Gaokao performance.
表 7.2006 年高考成绩达到最高门槛的简单对数回归。

Model 1 Model 2 Model 3 Model 4 Model 5
模型 1 模型 2 模型 3 模型 4 模型 5

Note: Standard errors are shown in parentheses.
注：括号内为标准误差。
${}^{\ast }p<0.10,{}^{\ast \ast }p<0.05,{}^{\ast \ast \ast }p<0.001$${}^{\ast }p<0.10,{}^{\ast \ast }p<0.05,{}^{\ast \ast \ast }p<0.001$^(**)p < 0.10,^(****)p < 0.05,^(******)p < 0.001{ }^{*} p<0.10,{ }^{* *} p<0.05,{ }^{* * *} p<0.001.

Model 2 introduces the indicator of parental educational level on Gaokao performance. When including parental educational level, the effect of socio-economic status tended to reduce. Parental educational level has a significant impact on students’ performance in the Gaokao. Students whose parents had completed higher education generally performed much better in the 2006 Gaokao than those whose parents did not. Model 3 demonstrates
模型 2 引入了父母教育水平对高考成绩的影响指标。当加入父母教育水平时，社会经济地位的影响趋于减弱。父母受教育程度对学生的高考成绩有显著影响。父母受过高等教育的学生在 2006 年高考中的成绩普遍比父母未受过高等教育的学生好得多。模型 3 表明
the gender difference in achieving the highest Gaokao performance. Male candidates were more likely to achieve the highest threshold in the 2006 Gaokao than were female candidates.
在取得高考最高成绩方面的性别差异。男性考生比女性考生更有可能在 2006 年高考中取得最高分。

Models 4 and 5 introduce a series of socio-demographic factors as additional predictors that influence academic performance in the Gaokao. Model 4 demonstrates a significant impact of the geographical origin on the Gaokao performance, even when socio-economic background is included. Specifically, students from cities and counties generally performed better than those from rural area in the 2006 Gaokao. When geographical origin was included, the effects of socio-economic characteristics significantly decreased and the impact of parental educational level also tended to reduce. Model 5 introduces types of secondary schooling, and demonstrates that key schools are strongly associated with higher academic performance on the Gaokao, even when all the other variables are included.
模型 4 和 5 引入了一系列社会人口因素，作为影响高考学业成绩的额外预测因素。模型 4 表明，即使将社会经济背景也包括在内，地域来源对高考成绩也有显著影响。具体而言，在 2006 年的高考中，来自城市和县城的学生的成绩普遍好于来自农村的学生。当加入地域来源时，社会经济特征的影响明显减弱，父母教育水平的影响也趋于减弱。模型 5 引入了中学教育类型，结果表明，即使包括所有其他变量，重点学校也与高考成绩的提高密切相关。

Students’ socio-economic status did not seem to have a consistent impact on academic performance when other socio-demographic factors were included. However, the impact of students’ parental educational level was consistently significant when socio-demographic factors were considered. Parental educational level and socio-demographic factors seemed to predict students’ academic performance in the Gaokao more strongly than students’ socio-economic backgrounds. This result seems to contrast with evidence presented in other contexts such as Japan, Korea, and Hong Kong, which suggests families from upper socio-economic status groups tend to invest in private tutoring or other extra-curricular activities to improve their children’s academic performance in the national entrance examinations (Bray 2007). A possible explanation for the relatively low impact of socio-economic status compared with parental education level might be that the selection to senior secondary schooling has already filtered out many students from poor rural areas or from lower-working-class backgrounds, but less so where the parents are highly educated. Students who participated in the Gaokao competition had already been pre-selected during the transition to senior secondary schools.
如果考虑到其他社会人口因素，学生的社会经济地位似乎对学习成绩没有一致的影响。然而，当考虑到社会人口因素时，学生父母的教育水平对学生学业成绩的影响一直很显著。与学生的社会经济背景相比，家长的教育水平和社会人口因素似乎更能预测学生的高考学业成绩。这一结果似乎与日本、韩国和香港等地的情况相反，这些国家的情况表明，社会经济地位较高的家庭倾向于投资于家教或其他课外活动，以提高子女在全国高考中的学业成绩（Bray，2007 年）。与父母的教育水平相比，社会经济地位的影响相对较小，一个可能的解释是，高中教育的选拔已经过滤掉了许多来自贫困农村地区或来自较低工人阶级背景的学生，但在父母受过高等教育的情况下，这种影响就较小。参加高考竞赛的学生在升入高中时已经经过了预选。

Table 8 reports the results of another series of simple logistic regression analyses of the log-odds of the second independent variable, university type. Here, the Tier One (985 institutions) and Tier Two (211 institutions) universities were coded as one, and the non-elite universities, including provincial vocational and technological universities, were coded as zero. Model 1 shows students from different socio-economic backgrounds and their destinations in different types of university. Students from professional backgrounds were much more likely to be accepted to elite universities than those from other social backgrounds. Students from leading cadre or executive manager families did not seem to have strong advantages in getting accepted into elite universities. Model 2 introduces the parental educational level along with socio-economic status. Students whose parents had higher education degrees
表 8 报告了对第二个自变量--大学类型--的对数的另一系列简单逻辑回归分析结果。在此，一级大学（985 所）和二级大学（211 所）被编码为 1，非精英大学（包括省属职业技术大学）被编码为 0。模型 1 显示了来自不同社会经济背景的学生及其在不同类型大学的去向。与其他社会背景的学生相比，专业背景的学生更容易被名牌大学录取。来自领导干部或行政经理家庭的学生似乎在被名牌大学录取方面并不具有很强的优势。模型 2 引入了父母的教育水平和社会经济地位。父母拥有高等教育学位的学生

Table 8. Simple logit regression of being enrolled into elite universities in the 2006 Gaokao.
表 8.2006 年高考被名牌大学录取的简单对数回归。