Digital Visual Analysis — Analysing Visual Materials in Digital Contexts

Standard visual analysis can be applied to any image — a painting, a photograph, a film still. Digital visual analysis specifically attends to the conditions of digital production, circulation, and consumption: platform affordances, algorithmic curation, engagement metrics, and the networked social contexts in which images acquire meaning. The analytical steps overlap substantially, but digital visual analysis adds a layer of platform-contextual interpretation that offline visual analysis does not require.

Corpus size depends on the research question and the analytic depth intended. In-depth semiotic or multimodal analysis of individual images may involve 20–100 purposively selected items. Broader pattern-oriented studies may examine hundreds or thousands of images using a combination of systematic sampling and close reading of representative cases. There is no universal minimum, but the corpus must be large enough to support the analytic claims being made.

This is contested and context-dependent. Images posted to fully public accounts on platforms like Twitter or public Instagram are often treated as available for research, especially when no individuals are identifiable or targeted. However, images from semi-public or community spaces, images of identifiable private individuals, and images from closed groups require more caution. Institutional ethics review boards vary in their guidance; researchers should document their ethical rationale and err toward anonymisation and consent when in doubt.

Yes — digital content is typically multimodal, and images on social media are usually accompanied by captions, hashtags, comments, and other textual elements. Multimodal discourse analysis or thematic analysis of accompanying text can be combined with digital visual analysis to provide a fuller account of how meaning is constructed across modes. The key is to be explicit about which analytic approach is applied to which data and why.

Qualitative data management software such as NVivo, ATLAS.ti, or Dedoose can store and code visual files alongside field notes. For corpus collection, tools like 4CAT, Zeeschuimer, or platform APIs are used. Multimodal annotation tools such as ELAN support detailed video analysis. None of these tools perform interpretation — that analytical work remains the researcher's responsibility.