著者

猪井 新一

出版者

一般社団法人大学英語教育学会

雑誌

大学英語教育学会紀要 (ISSN:02858673)

巻号頁・発行日

no.25, pp.39-52, 1994-09-07

---

The present study is aimed at investigating whether pronunciatlon questions on a written test are really valid as a means of evaluating a learner's actual pronunciation ability. A test of 20 questions on primary stress and another 20 questions on phoneme discrimination was administered to 44 college freshmen. The phoneme discrimination questions were made up of two subsections, each of which comprised 10 questions. The written test was followed by an oral version of the same questions. The data obtained from the written test and the oral test were analvzed in terms of the extent to which the answers were identical, and in terms of the extent to which scores on the written test were correlated with those on the oral test. As for primary stress, about 80 percent of the data was identical on the written and oral tests, and a significant correlation was observed between scores on the written and oral tests. As for phoneme discrimination, however, the overall agreement rate was about 10 percent lower than the one for primary stress, andno significant correlation was found on one of the two subsections. It can be concluded that primary stress quesions on a written test are fairly valid as a means of assessing a learner's pronunciation ability while the validity of certain types of phoneme discrimination questions on the written test is doubtful.