This study examines the relative contribution of vocabulary breadth and vocabulary depth to three different listening comprehension measures. One hundred and thirteen English majors were given VB and VD tests, and three listening comprehension tests. Based on three pairs of hierarchical multiple regression analyses, we found that the relative contribution of VB and VD varied across the three listening comprehension tests. Specifically, for the listening test with an expository text dictation to assess integrative skills, both VB and VD made a (...) unique positive contribution to comprehension, but this was greater in the case of depth. For the listening test involving narrative conversations to assess literal comprehension, neither VB nor VD could independently predict comprehension, whereas for the listening test that comprises expository passages to assess inferential comprehension, VD could separately predict comprehension but VB could not. These findings suggest that the relative contribution of VD and VB to listening comprehension may depend on how a listening test is constructed. Therefore, the findings will contribute to listening comprehension and vocabulary knowledge research, and vocabulary teaching and learning. (shrink)