The Visual Codebook pairs each coded segment of interview material with the image it refers to, so that a stretch of transcript and the work it describes can be read together rather than apart. Its analytical value lies in precision: when a passage that reads on the page as hesitation or vagueness is set beside the image the participant was discussing, the supposed vagueness often resolves into a precise non-verbal account of value.
The instrument was used within the coding, not after it, returning to coded segments with their images to test whether a code that named uncertainty was in fact naming a precision the researcher had failed to hear. What had looked like inarticulate participants were, in several cases, articulate participants working in a register the transcript alone could not carry. That reframing is what first suggested that the values category was not a gap in what participants knew but a gap in what a purely verbal transcript could record.
Twenty-five segments across five participants can be filtered by analytical category and participant, grouped by participant or by category, and expanded to show the full coding entry with image, theme, key quotes, parameter rationale, and an editable integration note recording what the instrument view changed in the analysis.
Thesis reference: section 3.5.1 (design), section 5.3.1 (contribution). Technical construction: Appendix C.