SeriesFusion
Science, curated & edited by AI
Paradigm Challenge  /  Economics

Graders for China’s big college entrance exam often ignore the rules to reward students who write essays with 'moral correctness.'

Despite strict psychometric protocols, this research found a 'shadow rubric' where graders' implicit moral biases override explicit linguistic criteria. This suggests that in high-stakes testing, cultural schemas of 'correct' thinking are often more powerful than the actual rules of the test.

Original Paper

Policy Mandates vs. Rater Cognition: Investigating Operational Validity in High-Stakes Chinese Writing Assessment

Lili Yu, Long Wu, Gang Huang, Pan Zhang, Fangfang Kai

SSRN  ·  6428284

In high-stakes assessment, writing score validity depends on raters consistently operationalizing official rubrics. Using the Gaokao (China’s National College Entrance Examination) as a case study, this research investigates the alignment—and dissonance—between policy mandates and practitioner cognition. Data from 1,241 senior high school Chinese teachers in Zhejiang reveals a profound "operational disconnect": while teachers show high nominal adherence to official standards, their practical app