Putonghua Proficiency Test Of Educational Statistics And Educational Measurement Research

Constitution of People抯 Republic China has stipulated, 揚utonghua should be popularized nationally?and State Three Departments also issued 揂 Regulation on Launching Putonghua Shuiping Ceshi?and 揂n Outline on Putonghua Shuiping Ceshi?(shortened as 揙utline?below). Putonghua Shuiping Ceshi (shorten as PSC below) is a main component of popularizing Putonghua and an important measure to develop Putonghua into a scientific, standardized and systematic stage. PSC, at present, is being popularized nationally whereas Yunnan province has achieved some accomplishments in the past ten years thanks to an earlier beginning. However, there is still a lack of scientific and accurate standard due to a deficiency in a support from educative statistics and measurement in theory, content, quantity as well as methodology. So in PSC 搘hat?and 揾ow much?should be examined in 搘hat?standardized way(s) is still a vital issue awaiting finther exploration and development. As a matter of fact, PSC, as a kind of language test, is characterized by widespread objects and special implementation, in which only with the aid educative statistics and measurement in theory and methodology, rather than from the experience, can accurate data and quota be acquired and judged and evaluated with more scientism and accuracy in a more extensive field. The object of PSC, unlike such in the physical measurements as weight, length and temperature that can be directly acquired with some precise instruments, is required to measure the language capability, a mental expression of examinee抯 language, intelligence, capability and mental activity. Therefore, we should back on the traditional Classical Test Theory while absorb more fresh ideas in PSC measurements in order to measure the Putonghua proficiency of examinee in a more objective and scientific way. Based on these characteristics the present paper is thoroughly embedded in a psychological measurement --- the Generalizability Theory.In the present paper, a statistical analysis is speculated and demonstrated on 2,646,000 data of 1,300 random test papers from five different-dialect representative districts in an attempt to solve the problems above. The results are shown as: (1) As to 搘hat?should be examined, PSC mainly focuses on the examinee抯 six aspects as general proficiency level, standardization, fluency, norm of Putonghua, and expression ability and style (talking way and style) of the examinee. (2) 揌ow much?should be examined is much determined by the varieties in validity required. So the interactive effect of different districts, nationalities, items and examiners were analyzed in details: when there were 7 examiners while 4 examining items, correlation between the examined results and the examinee抯 Putonghua capability reaches 0.9; when 2 examiners and 4 items, correlation is still 0.8691. Thus, it can be concluded that the four items in the present PSC can be measured in a rather precise way. That is, generally speaking, 揾ow much?should be examined can be responded in the present four items, single word, vocabulary, reading aloud and speaking. (3) As to 揾ow?to examine, such conclusion can also be drawn that only 2 examiners with 4 examining items can reflect the examinee抯 proficiency in a very accurate way. At the same time, it is suggested that the reliability in cooperation between different examiners and examining items is corresponded with the different companies according to the expected accuracy.