Four samples were chosen by systematic sampling from the range-finders between 1987 and 2003 to test the range-finders’ representativeness of the scoring criteria of CET-4 compositions. The samples were analyzed from...Four samples were chosen by systematic sampling from the range-finders between 1987 and 2003 to test the range-finders’ representativeness of the scoring criteria of CET-4 compositions. The samples were analyzed from the perspectives of transitivity, modality, thematic structure and cohesion. Fisher's Exact Test from the SAS system was used to compare two-sidedly the correct rate and the constituent ratio of these functional usages between two neighboring levels. Statistical results indicate that it is the quality, not the quantity of these functional usages that distinguishes two neighboring levels. The range-finders can basically represent the abstract scoring criteria as far as the four functional systems are concerned. However, the validity of CET-4 writing needs to be improved.展开更多
基金This paper is one of the outcomes of the university project ( XB200349)
文摘Four samples were chosen by systematic sampling from the range-finders between 1987 and 2003 to test the range-finders’ representativeness of the scoring criteria of CET-4 compositions. The samples were analyzed from the perspectives of transitivity, modality, thematic structure and cohesion. Fisher's Exact Test from the SAS system was used to compare two-sidedly the correct rate and the constituent ratio of these functional usages between two neighboring levels. Statistical results indicate that it is the quality, not the quantity of these functional usages that distinguishes two neighboring levels. The range-finders can basically represent the abstract scoring criteria as far as the four functional systems are concerned. However, the validity of CET-4 writing needs to be improved.
