The Equivalence of Direct and Semi-Direct Speaking Tests documents a comparability study of direct (face-to-face) and semi-direct (language laboratory) versions of the speaking component of the access test: an English language test designed in the 1990s as part of the selection process for immigration to Australia. In the study the issue of test equivalence is explored using a range of quantitative and qualitative evidence including test scores, test taker language output and feedback from various 'players' in the test process including members of the development team, test candidates, interviewers and raters. The findings have important implications for the use of direct and semi-direct speaking tests in various high-stakes contexts such as immigration and university entrance. As such this will be of interest to policy-makers and administrators as well as language teachers and language testing researchers.