THE DIRECT TESTING OF ORAL SKILLS IN UNIVERSITY FOREIGN-LANGUAGE TEACHING

In the past, the validity and reliability of oral examinations have been questioned, although more recent studies of reliability have tended to show that the problems might not be as great as was once imagined. The present article focusses on some practical problems that are said to accompany oral examinations, in particular the claim that a short oral examination does not provide an adequate basis for the assessment of oral skills because of the small sample of foreign language produced by the examinee. Samples from 15 minute oral examinations were compared with those produced by the same students in 3 hour written examinations, and it was found that the difference in size was rather small, although the variance in the size of the oral samples was considerably greater than in the written samples. The results have implications for the use and assessment of oral examinations.