Skip to content

We lack serious training in research methods choice and selection

As most of you who follow me on Twitter or read my blog probably know, I have been teaching research methods courses for a while now (mostly at the Masters and PhD levels). Thus I have been reflecting on issues related to training and teaching research methods. This morning, something really hit me as I was preparing an asynchronous lecture:

We lack serious training in research methods choice and selection.

We keep saying “choose the method according to the research question”. But this requires that the student KNOW multiple methods, for data collection AND data analysis. Moreover, developing criteria for method choice and selection requires students to learn heuristics that can help them build decision trees and models of the type
“IF research question 1
THEN (data collection method 1 AND data analysis 1)
ELSE search RM”

This is why research methods training is very hard, and why it’s imperative that whoever teaches methods goes over the details of multiple methodological and empirical strategies, using case studies: so students can see for themselves how researchers make methodological choices. I choose ethnography as a core methodological strategy because it helps me understand the lives of individuals facing water insecurity, toilet insecurity, and those who work in the informal waste sectors. I know how to use many other methods, but I make an actual choice/decision.

My research output in the past couple of months

One of the reasons why I assign multiple empirical analyses in my courses is that I want my students to learn HOW different researchers and research teams have made decisions on which methods for data collection, analysis and presentation they used. Which heuristics did they use? I was thinking of teaching an advanced course on ethnography, and realized that one of the ways in which I have learned how to improve my own ethnographic work is by reading book-length ethnographies. I have read many! I’ve learned about the decisions ethnographers made.

We need to change how we teach research methods. We really do. We need to seriously engage with the challenge of teaching method choice and selection. I believe one of the key gaps in methods training is method choice heuristics. How do I know how to choose a method to answer a specific question? How do I make a choice that gives me the answers I need? These are key questions that faculty teaching research methods need to teach students so they can, in turn, answer them.

You can share this blog post on the following social networks by clicking on their icon.

Posted in academia, research, research methods, teaching.

2 Responses

Stay in touch with the conversation, subscribe to the RSS feed for comments on this post.

  1. Nancy Augustine says

    I’ve been thinking about your post on the need to teach research methods selection. I wholeheartedly agree. I teach introductory research methods and capstone, and I haven’t figured out an effective way to develop research method selection.

    You allude to an important obstacle, that students have not been exposed to scholars’ thought processes in selecting a method. Some journal articles address it, but the author is speaking to a knowledgeable audience rather than beginners.

    Because our research methods course is coupled with introductory statistics, many of our students come to the course thinking that it’s something that they have to endure. Public administration and publicy policy do not attract students who are inclined to quantitative data analysis. We do very little with qualitative methods.

    I’m thinking about focusing on a single topic area during the semester and having students work through how the different methods allow them to explore different questions. For our students, a heuristic may be too advanced. Has anybody come up with other approaches?

  2. Gibran Vita says

    This is exactly what I am trying to do in my M.Sc. course. My approach is that, through small exercises and readings, I expose students to a “buffet” of about 15 (Qual & Quant) methods and frameworks. Then they have to write a research design of choice, with emphasis on methods.

    Also, I encourage them to characterize uncertainty on their method of choice and elaborate on how complementing or mixing with other of the studied methods would reduce that uncertainty.

    I think your idea sounds good in that sense, Nancy. Another approach that I have is that I make them compose a table of all methods studied since day 1. This table summarizes methods’ scope, their object of study, their advantages and disadvantages, type of data inputs, archetypical research questions they might answer, etc. My idea was that students can build and have this table at hand to navigate and choose more consciously.

    While I do get surprised about the quality of some of the assignments, I find a very prevalent issue which is that the methods selected do not really match the research question! Meaning, the data or insights generated through the method of choice wouldn’t answer the question. This is strange as all throughout the text and exam I have nudges encouraging them to double check for this matching and to iterate or re-formulate if necessary. This was the first course run so I will iterate.

    As you well say, a lot of practice in empirical research works well for this. What I am thinking is to have a workshop focused on peer 2 peer scrutiny of whether the R.Q. really match the method and vice versa, so hopefully everyone will be more sensitive to this and I can see different results in the exam, preparation for thesis writing.

    Even further, I also have a model on connecting epistemology to methods. This is because most STEM people never have the chance to reflect on this. In my opinion, not having this epistemic diversity in mind hinders inter/trans disciplinary work.

    Good reflections, Raul!

Some HTML is OK

or, reply to this post via trackback.