Associate Professor Sama Low-Choy
This session addresses the challenges of integrating mixed methods. We consider the proposition of Maxwell et al (2015:223) that: "The type of design, and the paradigm views of the researchers, are less important for integration than the ability to view the results using different mental models or "lenses." Direct engagement of the researcher(s) with both types of data, and ongoing interaction between quantitative and qualitative researchers, facilitates integration, as does systematically developing and testing conclusions using both types of data."
As a basis for examining integration of MM, this talk considers a collaborative project where high-level experts were interviewed about the risks of data linkage, involving social science data sourced from government. The methodology tightly connected thematic analysis, identifying themes in the transcripts from interviews, with a cluster analysis describing how the themes arise together. This combination of qual and quant research methods is a natural pairing, despite its relatively recent emergence. This talk explores whether integration of the qual and quant components is enhanced by different theoretical perspectives on the qual, quant or mixed methodologies.
From a practical and methodological perspective, this presentation also highlights some of the dangers of adopting new methods, and some steps that could be taken to avoid these. In this case study, clustering is a very new addition to textual analysis software. So, in comparison to other software dedicated to clustering (in bioinformatics, biological sciences, or machine learning packages), the software used has limitations regarding: support for new users, functionality specific to choice of algorithm and their settings, transferability of data, and transparency. This exemplifies issues encountered more generally by any researcher adopting new technology and/or methods new to them. We describe particular challenges in a mixed methods arena, either for those new to the quantitative method (here clustering), or new to the qualitative method (here thematic analysis).
Format; This workshop will be presented online, with some elements of learner-led education. Participants will be invited to email the presenter with specific questions that could help guide or exemplify content presented. An online forum will be opened a week prior to enable participants to share their preparatory reading. Small group exercises are designed to help digest ideas.
Pre-requisite and related workshops: You will find it easier to follow if you have attended any (or all of) the five workshops in the mixed methods (MM) series, presented via RED: (1) Introducing MM, (2) Refining and (3) Visualising the Conceptual Framework (CF), (4) Assembling MM, for surveys and interviews, and (5) Critical Reading & Writing of MM Research.
Reading: Maxwell, Joseph, Margaret Chmiel and Sylvia E. Rogers (2015) Designing Integration in Multimethod and Mixed Methods Research, Chapter 12, eds Hesse-Biber, S. & Johnson, N., The Oxford handbook of multimethod and mixed methods research inquiry, Oxford University Press, pp223-239.
Reference: Rose, J., Low-Choy, S., Katz, I., Homel, R. (under review) "Strategic data linkage to improve the wellbeing of vulnerable individuals: Reflections of experts".