Scholarlink

<p>This study investigates how China’s college admission system, including within-college major assignment mechanisms—<i>Major-level Parallel Mechanism</i> (MPM), <i>Major-level Immediate Acceptance</i> (MIA), and <i>Major-level Score Deduction Mechanism</i> (MSD)—shape student-major mismatch and its consequences for undergraduate development. Leveraging National College Admission Administrative Data (2005–2011) and a national undergraduate survey data (2014), this study employs Double/Debiased Machine Learning (DDML) based Instrumental Variable (IV) approach to estimate causal effects. Results reveal that students (compliers) who have been admitted under <i>Parallel Mechanism</i> (PM) and MIA interactively had dramatically higher possibility to be assigned an unpreferred major. This systemically induced major mismatch leads to a significant Local Average Treatment Effect (LATE)—declining academic performance, reducing research engagement, and lowering major identity without changing the overall career preparedness. The findings underscore the negative impact of misalignment between the application incentives and matching rules, advocate for reducing matching steps in China’s college admission system, phasing out the major adjustment policy or allowing post-enrollment unrestricted major switch under PM. This study advances matching theory in education and offers actionable reforms for centralized admission systems globally.</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

College admission mechanisms, major mismatch, and undergraduate development in China

  • 期刊论文

  • 发表日期：

    2026年4月7日

查看全文
  • Jin Yang

摘要

This study investigates how China’s college admission system, including within-college major assignment mechanisms—Major-level Parallel Mechanism (MPM), Major-level Immediate Acceptance (MIA), and Major-level Score Deduction Mechanism (MSD)—shape student-major mismatch and its consequences for undergraduate development. Leveraging National College Admission Administrative Data (2005–2011) and a national undergraduate survey data (2014), this study employs Double/Debiased Machine Learning (DDML) based Instrumental Variable (IV) approach to estimate causal effects. Results reveal that students (compliers) who have been admitted under Parallel Mechanism (PM) and MIA interactively had dramatically higher possibility to be assigned an unpreferred major. This systemically induced major mismatch leads to a significant Local Average Treatment Effect (LATE)—declining academic performance, reducing research engagement, and lowering major identity without changing the overall career preparedness. The findings underscore the negative impact of misalignment between the application incentives and matching rules, advocate for reducing matching steps in China’s college admission system, phasing out the major adjustment policy or allowing post-enrollment unrestricted major switch under PM. This study advances matching theory in education and offers actionable reforms for centralized admission systems globally.