Lexical density in candidate output on direct and semi-direct versions of an oral proficiency test

This article examines the effects of test format and task type on candidate output in direct and semi-direct versions of the oral interaction subtest of the access: test (the Australian Assessment of Communicative English Skills). Specifically, it contrasts the degree of 'orality' versus 'literacy' characterizing the audiotaped samples collected for the study from the perspective of lexical density, which provides a measure of the relationship between lexical and grammatical items in spoken or written discourse. The findings are discussed in relation to the degree of interactiveness and other factors which appear to influence lexical density. Finally, the results are examined in relation to the validity and interchangeability of the two kinds of oral proficiency tests.