Seeing Qualitative Data

← datawithaplot.org STORY

Where this comes from. Qualitative researchers publish fewer figures than they could, and the ones they do publish are mostly matrices: that is the finding of Verdinelli and Scagnoli (2013), whose nine display types (matrix, network, flowchart, boxed display, modified Venn diagram, ladder, decision tree, taxonomy, and metaphorical display) anchor the interpretive half of this catalogue. The ways of ordering a matrix follow Miles, Huberman, and Saldaña; the text-level displays follow Henderson and Segal (2013) and, for the word tree, Wattenberg and Viégas (2008); the framework matrix follows Gale and colleagues (2013); and the count-based forms rest on what Onwuegbuzie and Dickinson (2008) call quantitizing, counting coded themes and charting the result. The question about defending a count answers a pitfall Gale and colleagues name: "13 out of 20 participants said X" means nothing when the sample was built for range rather than representation. The six questions, and the mapping of which display suits which answer, are the author's own, drawn from building A Right That Exists on Paper and Maturity Gap, which is where the "see it in use" links go. The mapping is a starting point meant to be argued with, not a verdict.