The application of DICTION to analyse qualitative data: A luxury brand perspective

Abstract

Orientation: The evolution of technology has given rise to numerous computer-aided programs for analysing textual data, shaping the advancement of qualitative research. Research purpose: This study aimed to demonstrate how some of the critique associated with text analysis programs, as well as the dilemmas experienced in conducting research in the social sciences (achieving either precision or depth in the results, but seldom both), may be overcome. Motivation for the study: The South African luxury consumer market is described as complex and heterogeneous. Through the application of the text analysis software DICTION, this study demonstrated how this software can be applied to analyse in-depth interviews, and gain not only unique behavioural insights into a complex consumer group, but also produce quantitative output that allows for the development of distinct customer market segments. Research design, method and approach: A qualitative approach was adopted to collect the data, while the analysis using the text analysis program DICTION produced both qualitative insights and quantitative scores. The latter were used to perform a cluster analysis and develop unique customer segments. Practical/managerial implications: This study reveals how managers may swiftly obtain insights into lengthy transcripts of interview data, obtain quick answers about consumers’ behaviour and develop consumer market segments using the output obtained from the text analysis software DICTION. Contribution/value-add: The results demonstrate how both precision (statistical analysis) and richness (context from the transcripts) may be achieved in a research study, and provide a unique method to research, understand and segment a complex consumer market.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 1 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.