Unfulfilled Promise of Educational Meritocracy? Academic Ability and China’s Urban-Rural Gap in Access to Higher Education

Abstract

Abstract: With the rapid expansion of higher education, educational meritocracy has received both applause and skepticism among scholars, citizens, and policy-makers. Focusing on China’s urban-rural gap in college enrollment during the expansion, this study examines the differential effects of academic ability on urban and rural adolescents’ college enrollment. Using data from the China Family Panel Studies, the results show that the urban-rural gap in attending academic colleges is largest for adolescents who are at the middle ground of the distribution for academic ability, while the gap in vocational college enrollment is greatest for low-achieving adolescents. Compared to their urban counterparts, the positive effects of academic ability on academic college enrollment are stronger for high-achieving rural adolescents. It has little impact on the likelihood of college enrollment for low-achieving rural adolescents. The results suggest that the major source of China’s urban-rural gap in higher education is unequal access to vocational colleges. The findings provide important insights for understanding how structural, cultural, and policy factors perpetuate higher education inequality in China.

Acknowledgments

An earlier version of this paper was presented at the 2018 Annual Meeting of the International Chinese Sociological Association in Princeton, NJ, and was included as part of a working paper series of Princeton Research Network on Contemporary China. I wish to thank Simon Cheng, Scott Davies, Mary Fischer, Cindy Glovinsky, Jeremy Pais, Shi Pu, Xiaogang Wu, Yu Xie, and anonymous reviewers and editors from CSR for their helpful comments on earlier versions of the paper.

Notes

1 As suggested by Mijs (2016), since the meritocratic traits are influenced by nonmeritocratic factors, there is no neutral definition of merit. This study cannot rule out the possibility of measurement errors for academic ability and the presented results should be interpreted with caution.

2 Supplementary analyses using the sample of adolescents who were 14 to 18 years old in 2010 yield consistent patterns.

3 Following von Hippel (2007)’s recommendations, missing cases for the dependent variables were included in imputation equations but excluded from subsequent analyses.

4 Because of the data limitation, defining the postsecondary destination as the highest institution attended cannot capture students who transferred from vocational colleges to academic colleges and those who enrolled in colleges after 2014.

5 397 cases account for 27 percent of the initial sample, which may result in sample selection bias. In supplementary analyses, I estimate two separate probit models with sample selection to correct the potential selection bias that may be caused by the missing cases in academic and vocational college enrollment (Van de Ven and Van Praag 1981). The results yield consistent patterns.

6 In supplementary analyses, alternative model specification using hukou status to examine urban-rural differences yields similar results.

7 All reported regression estimates in this study are adjusted by the stratified multistage sampling design of the CFPS.

8 When comparing the coefficients of urban-rural status variable across differently specified logistic models, the interpretation of mediation effects of included control variables does not warrant the same interpretation of linear regression model (Karlson, Holm, and Breen 2012). Consequently, the KHB decomposition method is used to examine how the effect of urban-rural status on college enrollment shrinks upon the addition of control variables.

9 Additional analyses yield consistent patterns as all control variables are held at total sample mean values when calculating predicted probabilities and marginal effects, while the predicted probabilities for rural adolescents are slightly higher in these analyses and marginal effects remain stable.

Additional information

Notes on contributors

Angran Li

Angran Li ([email protected]) is a doctoral candidate in sociology at the University of Connecticut. His research interests include social stratification, sociology of education, urban and neighborhood sociology, higher education, quantitative methods, and Chinese studies. His recent empirical work has been published in Sociology of Education, Chinese Sociological Review, and Social Currents.