A typology of research questions about society

by Ingrid Robeyns on August 22, 2022

One of the things I really like about my job, is that I have been appointed on a chair with the explicit expectation to advance interdisciplinary collaborations between ethics and political philosophy on the one hand, and the social sciences (broadly defined) on the other. I’ve been co-teaching with historians, taught some courses that were open to students from the entire university, have been giving guest lectures to students in many other programs including economics, pharmacology, education, and geosciences; and I co-supervised a PhD-student in social work. I’ve written an interdisciplinary book on the capability approach, and have co-authored papers with scholars from various disciplines. So interdisciplinarity is deeply engrained in much of what I do professionally.

But while I love it enormously, interdisciplinary teaching and research is also often quite hard. One of the challanges I’ve encountered in practice, is that students as well as professors/researchers are not always able to recognise the many different kind of questions that we can ask about society, its rules, policies, social norms and structures, and other forms of institutions (broadly defined). This then leads to misunderstandings, frustrations, and much time that is lost trying to solve these. I think it would help us if we would better understand the many different types of research that scholars working on all those aspects of society are engaged in.

Thus, I’ve worked on a typology of different types of questions that can be asked in this area, discussed it with a range of colleagues who are based in other disciplines, and came up with the following (paper can be read open access here):

  1. conceptual research
  2. descriptive research
  3. explanatory research
  4. interpretative research
  5. evaluative research
  6. prescriptive research
  7. predictive research
  8. research developing visions
  9. methods, frameworks and other supportive research

As I write in the paper, I invite others (including students!) to revise and improve this typology, and am hoping it will be of help to anyone navigating an interdisciplinary context in the social sciences, and perhaps even for courses that discuss what interdisciplinarity entails.



Ben Wilbrink 08.22.22 at 6:46 pm

Historical research.


Ingrid Robeyns 08.22.22 at 7:21 pm

Ben Wilbrink – the typology is based on the “function” of the research – on what it tries to do, not on disciplines. I would suppose (though I am not a historian and don’t do historical reserach) in their work historians include a combination of the types of research in the presented typology – they describe, interpret, sometimes offer new concepts etc.

Or is there anything, in terms of such analytical/’functional’ categories, that historians do that you find missing?


nastywoman 08.22.22 at 8:32 pm

and the most fun you can have with your clothes on – is ‘cross sectional’ and ‘flexible’ research – in order to collect data that may help develop a new theory about a process or phenomenon called ‘Inductive’ and then the Deductive – theory-testing.


M Caswell 08.22.22 at 9:34 pm

Do any of the disciplines include modes of inquiry and reflection that aren’t research? If so, that might be a source of “misunderstandings.”


Matt 08.23.22 at 12:03 am

This looks very nice, Ingrid. I’ve only very briefly looked at the paper itself, but I wonder how or where you’d see “critical theory” of the Frankfurt school type fitting in – work that is supposed to be “essentially reflective” but both give knowledge and lead to liberation. Do you see it as a hybrid of some of the above? (I don’t think that’s how those doing such work would see it, but I’m an outsider to the tradition.) Or, is it not included for some other reasons?

(On a different note, are classes not normally open to any student at the university? In my experience in the US, there are cases where prerequisits would make it hard for many students not working specifically in a particular discipline to take a class, and sometimes cases where students from a particular school get “first shot” as seats in a class, but the default is that, for most introductory classes, and even for uper division ones w/o specific prerequisits, any student could take them. Is that not the case there?)


Ingrid Robeyns 08.23.22 at 6:02 am

Matt @ 5 – the method I’ve used for making this typology is like we often proceed in political philosophy – you read around in the literature, use your own brain, and then sollicit comments from a wide range of people to revise what you’ve done. For this paper, there was also years of teaching involved in which I saw a need for this kind of clarification, as well as interdisciplinary collaborations. When I had a draft I invited examples and comments from colleagues in various other disciplines. One of them, my colleague Joel Anderson, does have expertise on critical theory. So I take it that he must have thought, as you hint at, that critical theory can be seen as a hybrid of some of the other categories. But perhaps Joel knew how little time I had to finish that paper and just wanted to be supportive and nice to me :) – so curious to hear what other critical theorist think after they’ve read the paper.

What’s important is that the categories are indeed “analytical” – and in most pieces of research we combine different of these types of research (for example, the only ones that I know who do descriptive research without any interpretation or explanation at all, are the statistical offices; scholars typically start with descriptive research as the first phase, to then move to something else).


nastywoman 08.23.22 at 6:37 am

‘Deductive research approach –
When conducting deductive research, you always start with a theory (the result of inductive research). Reasoning deductively means testing these theories. If there is no theory yet, you cannot conduct deductive research.
The deductive research approach consists of four stages:
Start with an existing theory (and create a problem statement)
All Right-Wingers are ‘trump’
(The Worlds New Word for: Utmost Racist Science Denying Stupid)

Formulate a falsifiable hypothesis based on existing theory
If people are Right-Wingers, then they will always be ‘trump’
Collect data to test the hypothesis
Collect data of Right-Wingers
Test all Right-Wingers you can test for being ‘trump’
Study all Right-Wingers for being ‘trump’
Analyze and test the data

5 out of 100 Right-Wingers are not ‘trump’
Decide whether you can reject the null hypothesis
5 out of 100 Right-Wingers are not ‘trump’ = reject hypothesis
10 out of 20 aren’t ‘trump’ = reject hypothesis
All Right-Wingers depend on being ‘trump’ = support hypothesis

Limitations of a deductive approach
The conclusions of deductive reasoning can only be true if all the premises set in the inductive study are true and the terms are clear.

All Right-Wingers are ‘trump’ (premise)
Trump is ‘trump’ (premise)
Trump is ’trump’ (conclusion)
Based on the premises we have, the conclusion must be true. However, if the first premise turns out to be false, the conclusion that Trump is ‘trump’ cannot be relied upon.


albert 08.24.22 at 8:40 pm

Ingrid, I really like this typology and the first part of the paper (all that I’ve read so far). Are you aiming to build off of established debates or ways of teaching social research methods or are you imagining what the best way to describe forms of social research would be if we started from basics? I teach in an applied social science department with faculty who have PhDs in all sorts of fields and our methods courses therefore also need to (but don’t currently) function as philosophy of social science courses. Do you know of other work of this type that would be accessible to advanced undergraduates/beginning masters students?


Tracy 08.26.22 at 4:36 pm

Ingird, I applaud the approach here but I think it is overlooking the actual difficult;ty with interdisciplinary work. I headed a task force on one to the interdisciplinary core courses at our college for years. I enjoyed the courses but, as I suspect you have found, it is hard to get faculty to teach them. As you say, they are hard to teach well, mainly because they are so difficult to plan. But I don’t think this is because of the different research styles used by disciplines; it’s because they are formed around studying different things. They have different epistemes and that dictates and limits the kinds of knowledge they develop and how they do it. The difficulty we ran into is that a) many faculty didn’t want to learn new epistemes and b) felt they had shortchanged both themselves and the students by pretending to expound on knowledge they didn’t have. I always sympathized with this view; I was uncomfortable myself in the courses I taught, even after years of work. Having to meld research styles was part of this, but I could usually understand what other people were doing in terms of methods. My problem – and the problem of other scholars in our group – was that I was constantly on ground that felt unsteady. In one way, that was the whole point of the endeavor; we wanted the students to develop thinking across disciplinary silos and to embrace and deal with uncertainly creatively. But, to be frank, that was not what a majority of our students wanted and they felt as uncomfortable with our work as we did. Reading this experience over to research efforts, imho, leads to the same problems. Well, enough.


David Stein 08.28.22 at 5:48 pm

Philosopher Robert Brandom has characterized the pivotal idea of Kant (and modernity in general) as the discovery that “intentionality is normative.” In this light, derogating “presentism” takes on a distinctly presentist character.


John Quiggin 08.28.22 at 8:07 pm

@Tracy Australian undergraduate degrees are mostly tightly focused on a single discipline (to the extent that they aren’t purely vocational), but there has been a shift to some interdisciplinary courses, such as Politics, Philosophy and Economics (not sure how what we do relates to the Oxbridge model).

I teach in this program (I’m mainly a researcher, and this is my contribution to the teaching program). It’s true that there isn’t a lot of eagerness among others to teach in this course but I see this arising more from the result of disciplinary incentives towards specialisation than fundamental epistemic problems. After all, we expect the students to cope with a bunch of different disciplinary perspectives, so it ought not to be that challenging for academics. It’s more, I think, that people feel more comfortable in their own (sub)fields.

Of course, these are adjacent disciplines. It’s much harder to integrate natural sciences with PPE.


Tracy 08.29.22 at 5:44 pm

@John Quiggin We never tried to integrate either natural science or foreign language into the interdisciplinary core. The natural scientists wouldn’t hear of it and we couldn’t figure out how to do it. Same with, say, learning Korean in the language program. Like your place we end up trying to mix the social sciences and humanities.
I agree that specialization. has a great deal to do with the reluctance to be involved in interdisciplinary courses. I think you have the relationship between students and academics backwards, however. The main problem seems to me to be that academics can’t figure out how to plan the programs and, as a consequence, neither can the students. What we needed up doing in the core courses I supervised was allowing each a academic to weigh the course toward their discipline and bring others in as well as they could. For instance, when I taught the course I (political scientist) focused on social and political change, but illustrated the civil rights movement through the change in American music. This strategy worked in the sense that it mad faculty easier to recruit, but I was dissatisfied overall. This is really hard to do, iow.

Comments on this entry are closed.