Mixed methods

A topic in research methodology

Published research is often described as being 'mixed methods'.

Many research studies draw upon multiple techniques of data collection and analysis – for example this is common (indeed the norm) in case study research.

The term ‘mixed methods’ is often used as a descriptor of research studies, and there is a literature exploring this notion.

“Unfortunately, there are least three senses in which the term mixed methods tends to be used:

• A study is described as mixed methods because it uses several different techniques to collect and analyse data – I will label this meaning as MM-T (for techniques);

• A study is described as mixed methods because it collects and analyses a both quantitative and qualitative data – I will label this meaning as MM-D (for data);

• A study is described as mixed methods because it includes elements of both main research paradigms … – I will label this meaning as MM-P (for paradigms)….”

Taber, 2013: 112

MM-T is in a sense a trivial meaning for mixed-methods as this is common in many studies that are better described as case study, grounded theory or ethnography, for example.

(Read about the use of multiple research techniques in studies)

MM-D may be challenging to do well, as usually qualitative and quantitative data have different purposes in a study so have to be 'mixed' in a principled way. However, again, there is nothing special about studies that include both types of data.

So the term mixed methods is usually most appropriate when it relates to MM-P studies. MM-P research may prove challenging:

"To address specifically the issue of mixing ideas associated with research paradigms, we have introduced a concept called commensurability validity or legitimation, which is 'the extent to which the meta-inferences made [in a mixed methods study] reflect a mixed worldview based on the cognitive process of Gestalt switching and integration' … Commensurability legitimation will not be possible for many researchers, and it is difficult to learn how to switch perspectives and create new perspectives, but we believe that it is possible and desired. The strong (or fully) mixed methods position, we argue, is developed only after explicit and systematic consideration of qualitative and quantitative perspectives."

Johnson, Onwuegbuzie, & Turner 2007: 127

The distinction here is between

(a) studies that are mixed methods in the sense that they employ a diversity of approaches which are individually based upon quite different paradigmatic assumptions (i.e., (ontological) about the nature of what is being researched, and (epsistemological) how knowledge of it may be obtained), as suggested in Johnson and colleagues work,

and

(b) other studies that also employ multiple techniques but where those techniques draw upon a common set of coherent underpinning assumptions.

it is recommended that the term mixed methods is reserved for the former case where it does useful work in conveying something significant about the nature of the study, rather than used for all studies that employ multiple techniques.

Sources cited:

• Johnson, R. B., Onwuegbuzie, A. J., & Turner, L. A. (2007). Toward a Definition of Mixed Methods Research. Journal of Mixed Methods Research, 1, 112-133.
• Taber, K. S. (2013). Classroom-based Research and Evidence-based Practice: An introduction (2nd ed.). London: Sage.