| Chapter 1: | Introduction: The Importance of External Examinations in Education |
|
|
The size of the test-taking population in Sweden is about that of a Beijing suburb. Moreover, the test designed for the Swedish student may have behind it as much as 50 times the resources as the test designed for the Chinese student. The appropriate system chosen for Sweden might be to have each teacher individually design and grade selection examinations. The most appropriate selection system for China would have to be more standardized and machine scorable, whether measured in the proportion of correct responses, or in performance descriptors such as letter grades A, B, C, and so forth. The key ingredient is to construct the best selection test given the level of available resources and expected accountability.
Since the Second World War, the technology of examination administration has changed radically in OECD countries, but in many parts of the former Soviet Union and other parts of the world, these technologies have not kept pace. Today, as the chapter on Russia points out, the new Unified State Examination (USE) is in place. But until recently in the former Soviet Union, each faculty within each higher education institution would administer examinations independently. Even in mathematics and physics, examinations would primarily be delivered orally. They could only be taken in person at the faculty where they were designed. This system of selection was unfair, inefficient, and of low quality. It was unfair because those who could not travel to the examination site had less opportunity. The effect was to limit access to candidates who could afford to travel. It was inefficient because students must take a different examination for every institution to which they applied. Since they could not do this at a single sitting, they must wait for a next test-taking occasion. This could delay entry by a year or more. The opportunity cost for thousands of young people for this waiting period was thought by the state to be unimportant. It was of low quality because questions were often designed by elderly faculty members, isolated from recent changes in the labor market. They placed high weightings on out-of-date skills and designed tests the administration of which could not be standardized.