<p>This paper applies machine learning algorithms to measure inequality of opportunity in education. Compared with traditional approaches, machine learning methods for variable selection and model specification can better address the limitations of conventional estimation strategies. Using data from the China Education Panel Survey, we evaluate IOP in China’s basic education through both ex-ante and ex-post measures. The results indicate that IOP accounts for at least 20%–30% of the total inequality in educational achievement, with a more pronounced effect in urban areas than in rural areas. Among the determinants, school-level characteristics emerge as the primary driver of educational IOP, while family background plays a relatively smaller role. These findings suggest that ensuring equitable access to high-quality public education resources is essential for promoting fairness in education.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Inequality of opportunity in educational achievement in China: a machine learning approach

  • Qiuchuan Jiang,
  • Weiwei Zhao,
  • Kun Zhang

摘要

This paper applies machine learning algorithms to measure inequality of opportunity in education. Compared with traditional approaches, machine learning methods for variable selection and model specification can better address the limitations of conventional estimation strategies. Using data from the China Education Panel Survey, we evaluate IOP in China’s basic education through both ex-ante and ex-post measures. The results indicate that IOP accounts for at least 20%–30% of the total inequality in educational achievement, with a more pronounced effect in urban areas than in rural areas. Among the determinants, school-level characteristics emerge as the primary driver of educational IOP, while family background plays a relatively smaller role. These findings suggest that ensuring equitable access to high-quality public education resources is essential for promoting fairness in education.