Multimodal Podcast Learning to Build L2 Listening Self-Efficacy in Gen Z EFL Learners

Abstract

Gen Z learners are predominantly audio-visual, yet most educational podcasts remain audio-only, constraining their potential to support learners who depend on visual cues for comprehension. Addressing this limitation, the present mixed-methods study investigates how multimodal podcast learning combining audio, subtitles, and interactive transcripts enhances L2 listening self-efficacy among Gen Z EFL learners. Drawing on Bandura’s social cognitive theory and Mayer’s multimedia learning framework, sixty undergraduate students from Universitas Negeri Makassar participated in an eight-week intervention comparing multimodal and audio-only podcast instruction. Quantitative analyses indicated a significant improvement in listening self-efficacy for the multimodal group (M = 4.02, SD = 0.68) relative to the audio-only group (M = 3.25, SD = 0.70), t(58) = 7.91, p < .001, with a large effect size (Cohen’s d = 1.15). Qualitative findings identified three recurrent themes: strengthened confidence through repeated mastery experiences, increased motivation through authentic topics, and greater learner autonomy fostered by multimodal scaffolds. Collectively, the results demonstrate that multimodal podcast learning not only reinforces listening competence but also cultivates the psychological confidence required for self-regulated learning. The study advances theoretical and pedagogical understanding of how multimodality transforms podcast-based instruction into an inclusive and empowering practice for contemporary EFL education.