Musings on research, international development and other stuff

An idiot’s guide to research methods


I want to start by excusing the title – I am, of course, not implying that you are idiots. I have however been carrying out a lot of interviews recently and I noticed that many people – even some people working in the area of research and international development – get a bit muddled up about different types of research methods. In particular, quite a few people think that ‘quantitative’ research is synonymous with ‘experimental’ research. Not so. In fact, as I have discussed before, qualitative/quantitative and experimental/non-experimental are two different axes. For ease, I have created a wee table to demonstrate this below.

Sorry, I know that this is a deeply nerdy blog post and I promise to try to make the next one more entertaining. On the other hand, at least if you are going to have a job interview with a mean person like me who might ask you to describe different types of research, you now have a handy crib sheet! You are welcome.

research methods*Non-experimental approaches are also described by some (including DFID – see here) as ‘observational’. However (with thanks to @UCLanPsychology for pointing this out to me), be aware that this can cause confusion since some people use the term ‘observational’ to describe a way of collecting information (in contrast to other ways such as self-reporting or direct measurement) in both experimental and non-experimental approaches.


9 thoughts on “An idiot’s guide to research methods

  1. Qualitative research has its roots in interpretivism paradigm and quantitative research has roots in positivist approach. It is from this approach that we choose research instruments or you use the mixed methods approach.

  2. Years ago I came across an obscure essay by Henry Mintzberg (Mintzberg, Henry. (2005). Developing Theory About the Development of Theory. In Ken G. Smith and Michael A. Hitt (eds.) Great Minds in Management: the Theory of Process Development. New York: Oxford.: 7-8) that is both really entertaining (including taking on Karl Popper) and very helpful in sorting out terminology issues in research methods:

    “… let me try to clarify another confusion, the use of the terms ‘quantative’ [sic] and ‘qualitative’ when we mean ‘deductive’ and ‘inductive’. It is as if all deduction is quantative and all induction is qualitative. Not so. Theories can be assessed without numbers (even, dare I say, judgmentally – which, by the way, is what most seven point scales really amount to), just as numbers can be used to induce theories… This mix-up leaves the impression that ‘quantative’ research is somehow proper (or Propper) – i.e., ‘scientific’ – even if it contributes no insight, while qualitative research is something to be tolerated at best, and then only when exemplary. This is the double standard that pervades our academic journals to their terrible discredit. It also manifests itself destructively in doctoral courses that teach quantative methods (mostly statistics) as rites of passages. Those who cannot handle the fancy techniques cannot get the doctoral degree, even though there is all kinds of wonderful research with no numbers. Why not instead preclude from doctoral program students incapable of coming up with interesting ideas. Imagine that!”

  3. Pingback: An idiot's guide to research methods | kirstyev...

  4. Pingback: Dissertation101An idiot's guide to research methods | kirstyevidence » Dissertation101

  5. Pingback: 8 Research Methods Blogs - The Ultimate Keyword Research Website

  6. Let’s be careful about over emphasising the importance of control groups as the basis for differentiating between different methods. Exposure variation can also inform analysis. More generally, I guess that is a reminder the two by two matrices are more often the starting point for analysis than an end point…

    • Thanks for commenting James – but I am not sure I understand your point. Use of a control group is quite a common way of classifying different research methods – and is certainly an important distinction that we make in DFID. This is not the same as saying that presence or absense of a control group makes your method better or worse – but it is just a convenient and widely used way to categorise. On the two by two matrix – yes it is a simplification and we could probably find holes to pick in it.But on the other hand I am interested in making research methods easier to understand and this seems to be a way that helps people see the difference between two aspects – presence/absence of control group and qualitative versus quantitative. Having said that, I would be interested to know about other simple/introductory ways of describing research methods as I am quite sure there alternative and probably better ways.

Leave a Reply (go on, you know you want to...)

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s