Abstract
Multimodal listening resources are often recommended for vocational English for Specific Purposes (ESP) classrooms, but studies of learner preference typically treat vocational learners as a single group, even though ESP theory holds that instruction should match the specific needs of each learner group rather than a generic one. This study compared three vocational tracks, hospitality, tourism, and content creation, on their general preference for multimodal input and on the specific listening content and challenges they report. Data came from a bilingual questionnaire comprising 10 Likert-scale items and 4 open-ended items, completed by 76 Critical Listening students in a D3 English Study Program in East Java, Indonesia. Track membership was inferred from students' open-ended responses, yielding a single-track subsample of 47 for group comparison (hospitality n=12, tourism n=16, content creation n=19). Kruskal-Wallis tests found no statistically significant differences between tracks on any preference item, indicating that a general preference for multimodal input was not detectably different across tracks in this sample. Content analysis of the open-ended responses, however, showed clearly different content preferences and reported challenges: hospitality students favored staff-guest service scripts, tourism students favored tour-guiding and travel-narration material, and content-creation students favored vlogs, interviews, and social media content, with content-creation respondents also reporting more connectivity and distraction-related difficulties. Together, these findings suggest a practical distinction for ESP course design: the format of listening instruction can remain multimodal across a shared course, while the content it carries may still need to be differentiated by occupational track.