Mixed Methods Strategies for Handling Methodological Issues within Threshold Concept Features

David Reeping

The conference themes invite the community to reflect on the proliferation of work on threshold concepts and consider how the framework has been applied in practice. Accordingly, this paper presentation addresses the state of research designs in studying threshold concepts across the disciplines and offers strategies to engage their data in potentially unexpected ways.

Common critiques levied against the framework concern the loose definitions of the threshold concept features. The use of imprecise criteria in designs meant to identify ideas in a discipline fit to bear the title of “threshold concept” poses immediate threats to the research quality. Further complications emerge for researchers employing retrospective methods, i.e., asking participants to recall their learning experiences, where hindsight bias has been documented as a non-ignorable limitation. The several criteria combined with methodological issues in data collection place a considerable burden on researchers attempting to identify or describe threshold concepts, as the evidence required for a convincing argument demands a nontrivial research design.

Current trends in the modes of inquiry for interrogating disciplinary ideas as thresholds appear to show a preference for qualitative approaches, primarily using questionnaires and interviews as the methods of data collection. Those using multimethod or mixed methods approaches often employ a triangulation design but contain little conversation between the data. Integration strategies are generally ignored that could help the community grapple with the inherent difficulties in these designs.

This presentation will demonstrate strategies drawn from mixed methods research to aid scholars in creating research designs with a mixed methods approach – employing well-known methodologies like grounded theory and case studies. In particular, the focus will be placed on analytical strategies to facilitate integration between the different data strands in the study. Handling the definitional issues with threshold concepts will also be discussed through a mixed methods lens.