Mixed methods (mixing quantitative and qualitative research methods) and triangulation (using more than one method, or data set) are increasingly fashionable in research. As a more pragmatic approach, mixed methods can be seen as a third research paradigm (Johnson and Onwuegbuzie, 2004), a middle position which is essentially outcome driven.
While we have hinted at links between a particular method and a particular epistemological position, in practice ‘aligning a particular epistemology and paradigm with a particular methodology is not necessarily straightforward or helpful’ (Goodrick, 2011, p 11). Treating the two main approaches as totally separate, denying the possibilities for working back and forth between these two extremes is problematic according to Morgan (2007), who advocates a more pragmatic approach, examining ‘what people can do with the knowledge they produce and not on abstract arguments about the possibility or impossibility of generalizability’ (Morgan, 2007, p 72). Moving beyond attachment to one paradigm generally has’ greater power to convince a reader/examiner of the quality and value of your findings’ (Cooksey & Mc Donald, 2011, p 199). Mixed methods assumes that a methodology is chosen simply for its capacity to address research objectives. When using mixed methods is important to keep in mind
- how different data sets will co-operate
- different guiding assumption underlying each method chosen
Mixed methods involves collecting data in a simultaneous or sequential manner using methods drawn from both quantitative and qualitative traditions.
Mixed methods advantages
- Balances strengths and limitations of quantitative and qualitative
- Facilitates view from different perspectives
- Enhances credibility among a wider group
Combining quantitative and qualitative research allows (Bryman, 2004)
- ability to fill in gaps
- use of one methodology to facilitate other
- combining of static and procedural features
- gaining perspectives of researcher & participant
- role of theory in relation to research (deductive or inductive)
- epistemological orientation
- ontological orientation
Mixed methods requires
- Broader skill set
- Broader experience
- Time and resources
- Acceptance and understanding of mixed methods
Mixed methods research design
- What kinds of methods can be mixed?
- How can they be mixed?
- What is the rationale for mixing these particular methods?
- Are the methods being applied at the same time?
- Are the methods used in a chronological/sequential sequence?
- What is the relative weight given to each method? Is there a dominant method?
Adapted from Creswell & Plano-Clark (2007/2011), and Creswell (2009)
Example of concurrent triangulation design
Purpose: Examine adolescents’ attitudes towards recreation alcohol
Methods: Total of 563 adolescents were asked to complete a structured questionnaire on alcohol use. At the same time four focus group sessions with 8 to 10 young people in each were heldConcurrent embedded Design
Example: research in organisation carries out quantitative study of all employees, and at the same time qualitative interviews of cleaners
The qualitative research generally aims to follow up, to explore result from quantitative research done earlier. This approach tends to be chosen by researchers who favour a quantitative approach (Creswell, 2009). The research progresses in clear separate stages
- Weight is on qualitative
- Quantitative builds on result of qualitative
- Useful to develop an instrument, a model…
Example of sequential explanatory design
Purpose : examine and compare science classroom learning environments in Singapore and Australia from different perspectives.
Methods : Large scale quantitative questionnaire collecting data from 1081 students in Australia and 1879 students in Singapore
Data used as a springboard for further data collection involving interviews with participants, observations, and narrative stories
Triangulation was used to secure an in-depth understanding of the learning environment and to provide richness to the whole.
Involves the concurrent collection of quant and qual data.
Design issues in mixed method research
- What is the purpose of the research? – Who is the audience?
- What are your research questions?
- What methods will best help you answer those questions?
- Who do you need to ‘find out’ from? –Data sources
- What is the approximate sample size you will require to generate credible evidence?
- How will you analyse the data?
- How will validity/trustworthiness issues be addressed?
- What are your timelines? What skills/resource/ constraints/personal factors might enhance or limit the study?
Triangulation and mixed methods
Mixed methods involves more complex research. AND there is NO guarantee that research converges into one single story – multiple methods, multiple data sources, multiple analysts, multiple perspectives multiple or contradictory stories may need to be reconciled (Cooksey & Mc Donald, 2011). The aim is to understand areas of convergence and divergence for opening up a new understanding.
Consider the following vocabulary to explain how findings may link together (Based on Greene et al. 1989, p 259):
Triangulation: convergence, corroboration, correspondence or results from different methods.
Crystallisation: bring together findings
Complementarity: seeks elaboration, enhancement, illustration, clarification of the results from one method with the results from another
Development: seeks to use the results from one method to help develop or inform the other method
Initiation: seeks the discovery of paradox and contradiction, new perspectives frameworks, the recasting of questions or results from one method with questions or results from the other method
Expansion: seeks to extend the breadth and range of enquiry by using different methods for different inquiry components
Data Collection Methods
- Clinical/Medical data collection (quant)
- Document analysis (qual/quant)
- Diaries (qual/quant)
- Focus groups and group interviews (qual)
- Interviews (qual/quant)
- Observational strategies (qual/quant)
- Photographs and videos
- Q-sort (qual/quant)
- Scales – Quant- Semantic differential/Goal Attainment scales
- Survey (quant/qual)
- Tests (quant)
Bryman, A and Bell, E. (2011). Business Research Methods. Oxford University Press
Cooksey, Ray & McDonald, Gael (2011), Surviving and thriving in postgraduate research. Tilde University Press, Prahran
Cresswell, JW., & Plano-Clark, V.L. (2011). Designing and conducting Mixed Methods Research (2nd Ed). Sage: Los Angeles
Creswell, J. W. (2003). Qualitative, quantitative, and mixed methods approaches (2nd ed.). Thousand Oaks, CA: Sage.
Creswell, J.W. (2007). Qualitative inquiry and research design: choosing among five traditions, Thousand Oaks, CA: Sage.
Creswell, J.W. (2009). Research Design. Qualitative, quantitative and mixed methods approaches, Sage: Thousand Oaks.
Denzin, N & Lincoln, Y. (eds.) (1994). Handbook of Qualitative Research. Thousand Oaks: Sage.
Goodrick, D. (2011), Qualitative Research, Design, Analysis and Representation, unpublished notes from workshop presented at ACSPRI Canberra 2011
Greene, J. (2007). Mixed methods in social inquiry. San Francisco: Jossey Bass.
Guba, E.G and Lincoln, Y.S. (1994). ‘Competing paradigms in qualitative research’. In Denzin N.K. and Lincoln Y.S. (eds.) Handbook of Qualitative Research, 105-117.
Johnson and Onwuegbuzie, 2004 Mixed Methods Research: A Research Paradigm Whose Time Has Come Educational Researcher October 2004 33:14-26.
Onwuegbuzie, A.J. & Johnson, R.B. (2006). ‘The validity issue in mixed research’. Research in the schools, 13(1), 48-63
Minichiello, Victor & Kottler, Jeffrey A. (2010), Qualitative Journeys. Student and mentor experiences with research. Thousands oaks: Sage
Morgan, David L.(2007), ‘Paradigms Lost and Pragmatism Regained : Methodological Implications of Combining Qualitative and Quantitative Methods’. Journal of Mixed Methods Research 1/1, 48 – 7