I do wonder if I should doff my researcher cap more to statistics. Don’t get me wrong; I have an inherent prejudice of words over numbers when it comes to research. Yet a recent experience working on an education project at Edinburgh Napier University allowed me to realise the value that a statistical analysis can bring when using it alongside qualitative research – known as a mixed methods approach. Mixed methods is defined nicely by John Cresswell (2007) as:
a research design with philosophical assumptions as well as methods of inquiry. As a methodology, it involves philosophical assumptions that guide the direction of the collection and analysis of data and the mixture of qualitative and quantitative approaches in many phases in the research process. As a method, it focuses on collecting, analyzing, and mixing both quantitative and qualitative data in a single study or series of studies. Its central premise is that the use of quantitative and qualitative approaches in combination provides a better understanding of research problems than either approach alone. (p. 5)
Our small team was interested in investigating novel feedback initiatives for Higher Education students, and our research questions were very much led by our desire to understand what worked (I do wonder if I am a pragmatist at heart…). Our large sample size (710 students) meant it seemed almost rude not to invite statistical analysis into the research design. This, we felt, would help us see a pattern in our results – a trend – while interviews would allow us to probe into the ‘why’s’ behind these patterns.Whilst I and another qualitative researcher geared up the tape recorder for interviews and focus groups galore, the PI for the project flexed his statistical software in anticipation for some serious number crunching. Each of us were relieved the other was doing the ‘hard part’ of the research.
An in-depth explanation of this project and our rationale for choosing to work with a mixed methods approach can be found in our chapter in SAGE Research Methods Case “Mixed-Methods Research in Education: Exploring Students’ Response to a Focused Feedback Initiative“, but in this post I thought it would be useful to summarise some suggestions of what we, as individual researchers, learnt from working together in a mixed methods collaboration.
Working in a mixed methods research group can stimulate intellectual curiosity
It is unusual for a researcher to be ‘methodologically bilingual’: able to work equally comfortably with both quantitative and qualitative methods. Therefore, working on a research project with others who have strengths in different data collection techniques or analysis is a fantastic learning opportunity. For example, I managed to apply my statistical understanding that I had learnt in my MRes stats class to a real-world problem thanks to the PI’s love of explaining ANOVA tests.
Giving time to discuss your different viewpoints before you begin the research
As quantitative and qualitative methods can be argued to stem from different ontological and epistemological standpoints, it is important to discuss upfront how these will affect the study’s design. We made sure we were comfortable sharing our views because it led to a deeper and more equal understanding of each method’s strengths and weaknesses.
Considering publishing and funding opportunities before you start a mixed methods project
Some funders may be particularly keen to fund only quantitative or qualitative work, not a mix of the two. It is useful to be aware of any biases from particular research communities before applying for funding. Furthermore, writing up mixed methods projects for publications can be challenging as you have basically collected two different sets of data for one project and these need to be succinctly yet thoroughly represented within a strict word count. Therefore, it’s worth not only checking the word count required by different journals, but also making sure that the journal is ‘open’ to reviewing mixed methods research.
I would definitely work with mixed methods research again, but I still feel I’d need a statistical ‘expert’ to accompany me on the project. That in itself is a bonus as it’s much more fun working in a group than going solo (especially when the team exemplifies the Thesis Whisperer’s Circle of Niceness!).
Cresswell, J. (2007). Understanding mixed methods research. In J. W. Creswell, & V. L. Plano Clark (Eds.), Designing and conducting mixed methods research (pp. pp. 1–19).Thousand Oaks, CA: SAGE.