Comparing school-leaving tests across nine subjects in China and England: task granularity and scoring objectivity

Previous research suggests that knowledge in different subjects varies in terms of objectivity and ‘granularity’ (division of knowledge content into smaller chunks). It has also been suggested that subject objectivity and granularity vary across different educational systems, due to resource differences. This study measures knowledge objectivity and granularity manifested in school-leaving assessments across nine subjects and two educational systems—China and England. While previous studies have investigated assessments using vague operationalisations such as ‘essay’ or ‘short answer’ questions, this study presents a ratio variable to measure granularity: the implied time allocated to individual items. Objectivity is operationalised by coding the item scoring guides as consisting of entirely response content and thus being objective, or as including descriptive content and thus being subjective. It is found that, across both educational systems, STEM subjects have finer granularity and higher objectivity. China assesses with finer granularity and higher objectivity than England (except in mathematics, where granularity is similar and all items objective). England shows significantly greater variation in both variables. The findings offer empirical corroboration of claims about inter-subject differences across different national contexts. They suggest that finer granularity, higher objectivity, and less variation are adopted to cope with resource constraints.