A Semi-Automated Approach to the Content Analysis of Experience Narratives

Abstract

iScale will typically result in a wealth of experience narratives relating to different stages of products’ adoption. The qualitative analysis of these narrative is a labor intensive, and prone to researcher bias activity. This chapter proposes a semi-automated technique that aims at supporting the researcher in the content analysis of experience narratives. The technique combines traditional qualitative coding procedures (Strauss and Corbin, 1998) with computational approaches for assessing the semantic similarity between documents (Salton et al., 1975). This results in an iterative process of qualitative coding and visualization of insights which enables to move quickly between high-level generalized knowledge and concrete and idiosyncratic insights. The proposed approach was compared against a traditional vector-space approach for assessing the semantic similarity between documents, the Latent-Semantic Analysis (LSA), using a dataset of a study in chapter 4. Overall, the proposed approach was shown to perform substantially better than traditional LSA. However, interestingly enough, this was mainly rooted in the explicit modeling of relations between concepts and individual terms, and not in the restriction of the list of terms to the ones that concern particular phenomena of interest.KeywordsContent AnalysisSemantic SimilarityLatent Semantic AnalysisLatent ConceptAutomate ApproachThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.