Title
A Validation Study on the English language test in a Japanese Nationwide University Entrance Examination

Keywords: Japanese University Entrance Examinations, English language tests, JFSAT-English test, TOEFL Listening Comprehension Test

Authors
Akihiro Ito
Aichi Gakuin University, Japan

Bio Data
Akihiro Ito, PhD in Applied Linguistics (Hiroshima University, Japan) is currently an Associate Professor of applied linguistics at Aichi Gakuin University, Japan. His research interests include second language acquisition and language testing. In September 2001, he was given the JACET Award for the Most Promising New Scholar (Young Applied Linguist of the Year) for his article titled ‘Japanese EFL Learners’ Sensitivity to Configurational Distinction in English Realtivization’ in ITL Review of Applied Linguistics (Afdeling Toegepaste Linguistiek at Katholieke Universiteit Leuven, Belgie), 131&132, 11-33, 2001.

Abstract
The present study employs validation study on the English language test of the Japanese nationwide university entrance examination- the Joint First Achievement Test (JFSAT). Two studies are presented. The first examines the reliability and concurrent validity of the JFSAT-English test. The reliability was acceptable. Criterion validity was estimated by correlating the JFSAT-English test and English language ability measure (a carefully constructed cloze test) and was found to be satisfactory. The second study reports on a construct validation study on the test through internal correlation study. The JFSAT-English test was divided into five subtests. Examination of the correlation matrix indicated that the paper-pencil pronunciation test had low validity with almost no significant contribution to the total test score. It is argued that though the JFSAT-English test can work as a reliable and somewhat valid measure of English language ability, the paper-pencil pronunciation test should be eliminated and a listening comprehension test might be included as one of the subtests in the JFSAT-English test. The other subtests, however, showed satisfactory validity