The discovery of cumulative knowledge

Purpose - The purpose of this paper is to provide guidance for designing and generating cumulative knowledge based on qualitative research. Design/methodology/approach - The paper draws on the philosophy of science and specific examples of qualitative studies in accounting that have claimed a cumulative contribution to knowledge to develop a taxonomy of theoretically justified approaches to generating cumulative knowledge from qualitative research. Findings - The paper argues for a definition of cumulative knowledge that is inclusive of anti-realist research, i.e. knowledge is cumulative if it increases the extent and density of intertextual linkages in a field. It identifies the possibility of cumulative qualitative research based on extensions to the scope of the knowledge and the depth of the knowledge. Extensions to the scope of the knowledge may include expanding the time periods, context, and/or theoretical perspective used to explore a phenomenon. Extensions to the depth of the knowledge may include new empirical knowledge, methodological pluralism, theory elaboration, or analytic generalization. Individual studies can demonstrate their contribution to cumulative knowledge by locating their research within a typology/taxonomy that makes explicit the relationship of current research to past, and potential, research. Research limitations/implications - The taxonomy may be useful to qualitative researchers designing and reporting research that will have impact on the literature. Social implications - The increased use of research impact as an evaluation metric has the potential to handicap the development qualitative research which is often thought of as generating non-cumulative knowledge. The taxonomy and the strategies for establishing cumulative impact may provide a means for this approach to research to establish its importance as a contribution to knowledge. Originality/value - The concept of cumulative knowledge has not been systematically applied to research based on qualitative methods.