International language tests increasingly use group oral formats in both high- and low-stakes contexts. But how do test-takers with different characteristics interact in such formats? This book explores how test-takers with different extraversion and proficiency levels co-construct spoken interaction in groups. The impact of the two test-taker characteristics was examined across three task types and with two group sizes. Examination of group performances using both statistical and Conversation Analysis tools offered insights into how group conversation is co-constructed. By enhancing our understanding of the group oral test construct, the findings have useful implications for international language test development and enable appropriate choices regarding task type and group size.