When to go for mixed method approach?

Home Forums Methodspace discussion When to go for mixed method approach?

Viewing 9 posts - 1 through 9 (of 9 total)
  • Author
  • #3183
    Santosh Mishra

    It has been proven that a mixed method design may yield better data than alone either Qualitative or Quantitative. But could you tell when a researcher should opt for mixed method design?


    I think that in social science is quite common to use a mixed method design. Indeed objects analyzed are so complex that only a framework of analysis embracing more methods and means for data collection can offer a real and concrete understanding of the reality.


    There are a range of circumstances when it is justified to use mixed methods:


    1. When the phenomenon under study is complex, multifacetd and multilayered. It is unlikely that a mono method would be able to gather rich enough data to appropriately investigate. Related to this is when interventions are context specific.

    2. When the findings from one method needs further investigation.

    3. When different types of data are required (eg summative and formative evaluation).

    4. When one method is used to design another (eg focus group and then a survey)

    5. When corroborative evidence is required (triangulation)


    One thing – I think a quan and qual based definition, which is the common way the MM are defined, is too limiting. Mixing could also involve two quan methods. Issues such as the timing and nature of mixing, validation and quality apply as much in these designs as in qual/quan ones.


    The Journal of Mixed Methods is essential reading, as is the Handbook of Mixed Methods if your library has a copy.

    Katie Metzler



    This is a good chapter from Alan Bryman on mixed methods! You should be able to access this for free at the moment as we’re running a free trial of SRMO, our online research methods product. Take a look, you’ll find lots of good book chapters and journal articles on mixed methods!


    Happy browsing!


    sure , it is only when one method to begin with or simultenously even though, has one method results that inform the success of the other method can be well concieved and can yiled better results in the understanding of the research. But we face a problem of some of our supervisors who may not actually understand mixed method approaches suitable for a particular research do confuse more the researcher at the time of their defenses and agitate for either qulitative and or quantitative alone. To me ,the problem with any type of mixed method is the time of combining the results to give an interpretation of them becomes a problem that i always read about. I aso agree with Richard Griffin that in all cases say in triangulation requires combining the results  that corroborate,but this must be after a careful resolve I’m however not yet there but it is an i envisage.


    I think when you stand on pragmatism as your ontological stand and think that research problem is more significant than research methodology.


    Hi Kapil,  pragmatism has a bearing, it has a philosophy behind its application therefore a research problem defines the research question that informs the research methodology that can be either combining qual and quan or separately carry out a survey separately. The methodology will direct the expections of the results as related to the theory or theories that define or answer the research questions. So can u put more light on the above assertions you make so that the mixed methods fraternity can benefits for your understanding  of your statement above.



    Pragmatism certainly suggests that the starting points of research should be what is known, where doubt remains and what is the best means (methodology and methods) to address a research question. This is in contrast to a foundationalist position which fixes the methodology and methods depending on the paradigm’s ontological position (eg realism). Peirce said that foundationalism places blocks on the road of inquiry. Pragmatism is pluralist in its approach to investigation and rejects incommensurability which is why it is frequently twinned with mixed methods – although pragmatism can justify mono approaches.

    Paul Kiff

    Your question presupposes that mixed methods are always optional.


    you need mixed methods if you are trying to collect both qualuitative and quantitative data


    All quantitative surveys are usually and should be derived from prior qualitative assessment of what questions to ask and how to ask them


    All quantitative and qualitative analysis need interpretation and all interpretation is usually and should be assessed qualitatively and quantitatively


    It is a fallacy to imagine that research can be anything other than mixed method

Viewing 9 posts - 1 through 9 (of 9 total)
  • You must be logged in to reply to this topic.