The topic of this article is the C-Test — a relatively new addition to the existing battery of techniques, cloze in particular, which claim to measure general language proficiency. Recent trials of th C-Test in Oman on subjects at lower levels of attainment have produced encouraging results on all counts except discrimination. This paper reports an attempt to address the problem of poor discrimination by means of an experimental C-Test variant in which grammatically unmarked items were deleted to the left rather than to the customary right. The findings showed that the discrimination of the C-Test could indeed be enhanced by left-hand deletions, but for a negative rather than a positive reason. It is concluded that the C-Test, and possibly other testing techniques which have proved effective at intermediate level and above, may be less appropriate when applied to lower levels, and that caution should be exercised in their use.
BackgroundIn recent years the University of Duisburg has developed a reducedredundancy testing format known as the C-Test. The major publications are Raatz and Klein-Braley (1982), Klein-Braley and Raatz (1984), and Klein-Braley (1985). The C-Test is an adaptation of cloze requiring word completion, and it is intended by its authors to supersede cloze as a measure of general proficiency.