Enriching thematic analysis with clustering techniques: applying mixed analysis to interviews about big data linkage

Researchers in business increasingly face a deluge of textual data. Often, the challenge is to reveal patterns in what has been said. We describe how interview transcripts can be analysed both qualitatively via thematic analysis, to manually identify latent themes that recur in the text, then quantitatively via cluster analysis, to see which ideas tend to appear together or apart. This differs from automated content analysis that finds key words before clustering. The deeper insights obtained using mixed (qualitative and quantitative) analysis are demonstrated using our recent study, analysing interviews with leaders on linkage involving big data. We take advantage of new software functionality that seamlessly supports manual coding of themes, followed by clustering of selected themes, on one platform. Although easy to implement, options are currently limited or hidden. Thus, results are open to misinterpretation. We note benefits and dangers inherent in integrating thematic analysis with clustering.