This chapter reports on a mixed methods study that investigates the effect of the interlocutor’s native/nonnative-speakerness on the test takers’ production of repair practices. The study was conducted with 28 second-semester learners of Chinese participating in two classroom-based speaking tests with different student interlocutors, a peer and a native speaker, both of which were students. Using a conversation analytic approach, we identified nine practices of other-initiated repair (i.e., nine different ways of indicating non-understanding and asking for help) and five types of other-directed word searches (i.e., five ways of recruiting the coparticipant’s help in the face of a speaking problem). While we found that the students employed both other-directed word searches and other-initiated repair more frequently in the conversations with the native speaker, the analysis also showed a lot of individual variability in the production of repair across the two speaking tests. Even though not all test takers and native speakers react in the same way, the findings of the study indicate that a linguistically asymmetric test setting by including a native speaker as interlocutor may be most fruitful if the testing objective is to elicit repair practices from the test taker.