Numerically Aided Phenomenology: Procedures for Investigating Categories of Experience

Abstract

Complementarity between quantitative and qualitative methods often implies that qualitative methods are a step toward quantitative precision or that quantitative and qualitative methods provide mutually validating "triangulation." However, there also is unacknowledged quantification within the type of analytic induction that is considered pivotal in qualitative thinking. We attempt to justify this claim and present a form of phenomenological analysis that invokes numeric algorithms. Numerically aided phenomenology is a procedure for systematically describing categories (kinds, or types) of lived experience within a set of experiential narratives. In a comparative reading, recurrent meaning expressions are identified and paraphrased. Then judgments about their presence or absence are used to create matrices representing the profiles of meanings expressed in each narrative. Finally, cluster analytic algorithms are used to group these experiential narratives according to the similarities in their profiles of meaning expressions. In this way, categories of similar experiential narratives—and their distinctive attributes—can be identified. Rather than an essentialist conception of the qualities defining classes, in numerically aided phenomenology classes are defined by more-or-less invariant attributes, i.e., classes are formed such that members share a large number of expressed meanings, although no single meaning (or set thereof) is necessary or sufficient for class membership. URN: urn:nbn:de:0114-fqs0101153