The Effect of School Choice on Student Mathematical Achievement Score: The Case of China

• In Event: What Chinese Students Really Know and Can Do: Evidence Beyond the Data From PISA (Programme for International Student Assessment) 2009

Abstract

Shanghai-China not only performs on average significantly better than the rest of the countries/districts that participated in PISA 2009, the score variation is also among the lowest. Compared to an average of 7.6% of high-performing students across countries, Shanghai’s percentage is about twice this figure. About 14% of students in Shanghai-China attain the highest levels of proficiency in all three assessment subjects (OECD, 2010).
The effect of schooling on student achievement has launched three different, yet related domains of research: effective school research (Lee, 1995), school choice research (Hansen, Heckman, & Mullen, 2004), and social capital research (Cabonaro, 1998). Among others, school choice research focuses on differences between a student’s actual performance and the performance that would have been expected if that student had attended a given type of school (Raudenbush & Willms, 1995).
In China, school choice has been a major driving force for introducing the market mechanism into the schooling process. Because of the culture, competition for school choice among parents becomes fierce and bears extensive consequences on social equality and educational policy.
In the current study, the effect of school choice in China is investigated. Based on Rubin’s counterfactual model of causal inference (Rubin, 2000), exogenous factors of schooling, such as student SES and intellectual ability, will be controlled via propensity score stratification, in order to make valid inferences about the effect of key versus ordinary schools on students’ mathematical test scores. Two questions will be addressed: (1) Do key (higher ranked) schools do a better job than non-key schools in promoting student achievement? (2) If so, what are those central characteristics that effectively distinguish different key schools with various magnitudes of effects? Implications of such findings for school choice and educational policy will be discussed.

Authors

• xiangdong Yang, East China Normal University

• Zheng Ke, East China Normal Uniiversity

• Tingting Wang, East China Normal Uniiversity