Abstract

The communicative landscape of social media is fundamentally multimodal, with meaning co-constructed through the interaction of text, images, videos, and platform-specific affordances. Yet research methods often lag behind this complexity, with many studies reverting to text-centric approaches that limit our understanding of online discourse. This gap calls for integrative approaches that connect theory-driven annotation and analysis with scalable computational techniques, enabling the systematic study of multimodal meaning-making.

This talk follows this trajectory, moving from conceptual rationale to practical application. It begins by showing how theoretical perspectives from discourse and communication studies can inform methodological choices, including data sampling, unit definition, tool selection, and the principled design of annotation schemes and analytical models. Building on this foundation, the session illustrates how qualitative and computational tools can be combined to achieve both interpretive depth and analytic scalability. Examples will include the use of CAQDAS software for corpus-assisted multimodal discourse analysis, computer vision APIs for large-scale image tagging, and multidimensional analysis techniques for identifying co-occurring communicative features across modes.

The presentation also considers the ethical and practical challenges involved in collecting, cleaning, and managing multimodal social media data. By the end of the session, participants will be better equipped to design transparent, reproducible, and theoretically grounded workflows for their own research. The talk will be of interest to scholars in linguistics, communication, and related fields who wish to move beyond monomodal analysis and adopt scalable, theory-informed approaches to studying the multimodal dynamics of social media.

Speaker biography

Dr Yuze Sha received her PhD in Linguistics from Lancaster University, UK. Her research examines digital and public health communication, with a particular focus on how legitimacy is negotiated in contested online environments. She is also interested in methodological innovation in multimodal and corpus-based discourse studies. Her work has been published in leading peer-reviewed journals such as Linguistics Vanguard. Dr Sha has presented her research at major international conferences and invited workshops.