選択式クローズテストと本文の語数(第2言語習得論・英語学力論)

概要

論文の詳細を見る
This study examines what is the most appropriate number of words in the text for constructing the most effective M-C cloze test. The following tests were administered to college sophomores: CELT No.1 (Listening) and No.2 (Structure & Vocabulary) in 1990; 6 M-C cloze tests,2 of them with approximately 400-word texts, another 2 with approximately 500-word texts, and the other 2 with approximately 600-word texts from 1990 through 1991. Six surveys were conducted on the level of difficulty, the length of the text, and the testing time. Results indicate the following things : First, there is no outstanding preference for a certain number of words used in the text. Second, there was a moderate but not high correlation between an M-C cloze test,and CELT No.2. Third, there is almost no correlation between an M-C cloze test and a listening test. Fourth, the reliability of the M-C cloze test does not vary according to the number of words used in the text. What deserves special attention is that 3 out of 6 M-C cloze tests surpassed CELT No.2 in reliability. This study indicates that an M-C cloze test has greater potential as an effective means of measuring a language learner's overall language proficiency.
全国英語教育学会の論文