Making a Transdisciplinary Dissertation Work
The Problem with Transdisciplinary Projects
https://www.google.com/maps/d/u/0/edit?mid=14aiKL_KZOlrq4PrYctsXeNpr32RKC2f-
Traditionally, a good research paper and even a masters thesis or dissertation follows a fairly standard IMRAD (Introduction, Methods, Results and Discussion) format. But what happens when you are not interested in doing a standard project in one primary academic discipline? You are interested in combining ideas across multiple fields to do a project that attacks your problem of interest from several inter-related perspectives. This is the core of an interdisciplinary, or possibly, a transdisciplinary project. We can argue about the definition of a transdiscipline, but for shorthand, here I mean drawing on multiple academic disciplines, integrating theory and research methods from these, and combining this interdisciplinary academic effort with real-world application.
Many students get tripped up by these projects, and in particular get stuck on how to do the write up. Part of the issue is that the transdisciplinary problem space does not lend itself to linear thought, you are thinking about many inter-relationships between theory, experimental methods taken from different disciplines, and analysis techniques that may also differ across those disciplines. You may be trying to mix and match ideas across these different areas. It can feel very compelling, but also very confusing. And as a result, the first version of your dissertation introduction chapter confuses you, your chair, and your committee.
Lean on Linearity - despite the fact that you are a non-linear thinker.
The simple answer is that you need to actually *rely on the linearity* of the IMRAD format and the form of the dissertation document, but look at it as extensible. So instead of doing an introduction on one topic from one discipline, you need to think about how that topic would be looked at theoretically by each of the disciplines you are drawing from and write a mini theoretical introduction that is well detailed and well cited within *each* of those perspectives.
Using an example from one student I'm working with now, we put the following outline - in broadest terms - of his dissertation plan together yesterday at the whiteboard. He is interested in using a data set gathered to look at early warning signs of crisis in US military veterans. The lab is working on computational approaches to improve our ability to detect early warning signs. Thus, the student needs to be familiar with at least 4 major theoretical areas - crisis theory (broadly speaking), mental health crises (more specifically), ecological momentary assessment (the data collection strategy that was used), and machine learning and statistics (his primary academic discipline) + the lived experience of US military veterans and/or how his work will be applied in practical ways to help veterans (in ways that they will accept/adopt).
Diagram Your Dissertation
These different disciplinary focus areas are described here as a Venn diagram. While it is a simple representation, consider again that each area has its own theories, its own research methods, and may have distinct views on what qualifies as evidence. Thus, the central point of overlap in the diagram is worth serious consideration - it is not just a pretty picture. It is the confluence point of several distinct ways of thinking about the problem you are trying to attack as part of your dissertation from an academic perspective (Fig 1). However, I would again argue that the first figure is really representing an interdisciplinary space, not a transdisciplinary one.
Understand How Transdisciplinarity is Different
There are a few ways of looking at transdisciplinarity, one of which suggests that rather addressing the problem from several independent perspectives, the researcher must integrate these theories and methods into a new set of approaches that merge the key strategies into a single approach that addresses the problem space more comprehensively than any of the approaches on their own.
Another perspective on transdisciplinarity is that the findings must also be applicable not just in the laboratory, but also in the real world. In this example, the "real world" has to do with the culture and lived experience of US military veterans (Fig. 2). If a set of solutions (in this case likely a set of clinical decision support systems) does not take into account veteran culture, the chances of uptake of the intervention/innovation are small. As one of my colleagues reminds me, academics are often focused on the next research innovation, but real world uptake of solutions proposed by academia is also its own form of innovation -- one with much greater social impact.
Example of Dissertation Table of Contents
In practical terms, this might play out something like this for the actual dissertation TOC. Note that this is an abstraction, and imagine inserting the particulars of your own project instead of these areas of work:
CHAPTER 1 INTRODUCTION
Subsection 1.1: Crisis theory
Subsection 1.2: Mental health crises
Subsection 1.3: Veteran culture & lived experience
Subsection 1.4: Ecological Momentary Assessment
Subsection 1.5: Machine learning
CHAPTER 2: INTEGRATION AND INNOVATION
Subsection 2.1: Explores and integrates key features of all four disciplines covered in Chapter Subsection 2.2: Discusses innovative edge in the transdiciplinary space you have identified (the center of the Venn diagram)
-OR-
Subsection 2.2: Discusses innovative edge within your primary discipline that is informed by these other ways of thinking (i.e. where it the boundary of your primary discipline is touching all of the other disciplines involved).
CHAPTER 3: INTEGRATED RESEARCH GOALS, METHODS, INSTRUMENTS & EVIDENCE
Traditionally, a good research paper and even a masters thesis or dissertation follows a fairly standard IMRAD (Introduction, Methods, Results and Discussion) format. But what happens when you are not interested in doing a standard project in one primary academic discipline? You are interested in combining ideas across multiple fields to do a project that attacks your problem of interest from several inter-related perspectives. This is the core of an interdisciplinary, or possibly, a transdisciplinary project. We can argue about the definition of a transdiscipline, but for shorthand, here I mean drawing on multiple academic disciplines, integrating theory and research methods from these, and combining this interdisciplinary academic effort with real-world application.
Many students get tripped up by these projects, and in particular get stuck on how to do the write up. Part of the issue is that the transdisciplinary problem space does not lend itself to linear thought, you are thinking about many inter-relationships between theory, experimental methods taken from different disciplines, and analysis techniques that may also differ across those disciplines. You may be trying to mix and match ideas across these different areas. It can feel very compelling, but also very confusing. And as a result, the first version of your dissertation introduction chapter confuses you, your chair, and your committee.
Lean on Linearity - despite the fact that you are a non-linear thinker.
The simple answer is that you need to actually *rely on the linearity* of the IMRAD format and the form of the dissertation document, but look at it as extensible. So instead of doing an introduction on one topic from one discipline, you need to think about how that topic would be looked at theoretically by each of the disciplines you are drawing from and write a mini theoretical introduction that is well detailed and well cited within *each* of those perspectives.
Using an example from one student I'm working with now, we put the following outline - in broadest terms - of his dissertation plan together yesterday at the whiteboard. He is interested in using a data set gathered to look at early warning signs of crisis in US military veterans. The lab is working on computational approaches to improve our ability to detect early warning signs. Thus, the student needs to be familiar with at least 4 major theoretical areas - crisis theory (broadly speaking), mental health crises (more specifically), ecological momentary assessment (the data collection strategy that was used), and machine learning and statistics (his primary academic discipline) + the lived experience of US military veterans and/or how his work will be applied in practical ways to help veterans (in ways that they will accept/adopt).
Diagram Your Dissertation
These different disciplinary focus areas are described here as a Venn diagram. While it is a simple representation, consider again that each area has its own theories, its own research methods, and may have distinct views on what qualifies as evidence. Thus, the central point of overlap in the diagram is worth serious consideration - it is not just a pretty picture. It is the confluence point of several distinct ways of thinking about the problem you are trying to attack as part of your dissertation from an academic perspective (Fig 1). However, I would again argue that the first figure is really representing an interdisciplinary space, not a transdisciplinary one.
Understand How Transdisciplinarity is Different
There are a few ways of looking at transdisciplinarity, one of which suggests that rather addressing the problem from several independent perspectives, the researcher must integrate these theories and methods into a new set of approaches that merge the key strategies into a single approach that addresses the problem space more comprehensively than any of the approaches on their own.
Another perspective on transdisciplinarity is that the findings must also be applicable not just in the laboratory, but also in the real world. In this example, the "real world" has to do with the culture and lived experience of US military veterans (Fig. 2). If a set of solutions (in this case likely a set of clinical decision support systems) does not take into account veteran culture, the chances of uptake of the intervention/innovation are small. As one of my colleagues reminds me, academics are often focused on the next research innovation, but real world uptake of solutions proposed by academia is also its own form of innovation -- one with much greater social impact.
Example of Dissertation Table of Contents
In practical terms, this might play out something like this for the actual dissertation TOC. Note that this is an abstraction, and imagine inserting the particulars of your own project instead of these areas of work:
CHAPTER 1 INTRODUCTION
Subsection 1.1: Crisis theory
Subsection 1.2: Mental health crises
Subsection 1.3: Veteran culture & lived experience
Subsection 1.4: Ecological Momentary Assessment
Subsection 1.5: Machine learning
CHAPTER 2: INTEGRATION AND INNOVATION
Subsection 2.1: Explores and integrates key features of all four disciplines covered in Chapter Subsection 2.2: Discusses innovative edge in the transdiciplinary space you have identified (the center of the Venn diagram)
-OR-
Subsection 2.2: Discusses innovative edge within your primary discipline that is informed by these other ways of thinking (i.e. where it the boundary of your primary discipline is touching all of the other disciplines involved).
Subsection 2.3. Offers an integrative set of overarching, theoretically driven research questions.
My feedback to a student on formulating research questions:
You want to offer maybe 2 or 3 broad research questions here – these are not operationally defined aims – but the aims you present (in the next section) help answer these research questions. The research questions should really also explain why this research is important and why the results would offer important innovations – what is new knowledge that would be produced? This new knowledge may be specific ML strategies, it might be the statistical findings, it might be that we have a better understanding of the informational needs of veteran peer mentors, etc. Students often get stuck on the idea that they have to present some huge new finding – that is not the point. Keep in mind that science is almost always incremental. We are looking for small, but important wins in terms of producing new knowledge, and thus offering the world just a little bit deeper insight into the problem itself, but more importantly – how to solve the problem. And this brings up what the real problem is – it is that often veterans go into crises and know one even knows – they are home, alone, and end up so isolated from people who could help that they commit suicide, drink themselves to death, hit their wives, abuse their children, end up going to jail, etc.
CHAPTER 3: INTEGRATED RESEARCH GOALS, METHODS, INSTRUMENTS & EVIDENCE
Your hypotheses, aims or goals depending on the methods you will use, along with operational definitions for key variables fitting the innovations to your specific data and the constraints of those data. You will want to integrate strategies from more than one discipline in many cases, although it is possible that you will also have some research goals that are driven from a single disciplinary perspective to build to this more integrated approach.
ExampleOne example we recently came across was the desire of a student to not use a simple diagnostic cut score approach for a psychometric instrument that is typically used to provisionally categorize people into a binary diagnostic class - i.e. positive or negative for PTSD. This perspective is driven from the practical therapeutic and more abstract nosological considerations in psychology, and usually a cut-score is used. The student wanted to use a machine learning approach with three levels of severity to examine predictors for these severity levels. Rather than arbitrarily setting severity levels, a deeper examination of the PCL-5 literature offered some guidance on where to set these levels, and this approach lead to the student publishing an IEEE paper.
Depending on the primary discipline or requirements of integrating several approaches, it is possible that traditional hypothesis testing may not be the most fitting approach, and the effort may instead be structured as a set research goals or aims that can be evaluated using the evidentiary rules of disciplines that are driven from a non-hypothesis based view of scholarly insight. The key here is picking a research strategy that is appropriate for your questions and will generate compelling data from an integrated perspective.
CHAPTER 4: METHODS
CHAPTER 5: RESULTS
CHAPTER 6: DISCUSSION
Should offer theoretical predictions that are based on the transdisciplinary space, not just the primary discipline. Should summarize methodological innovations that are distinct from those of any of the primary disciplines
CHAPTER 5: RESULTS
CHAPTER 6: DISCUSSION
Should offer theoretical predictions that are based on the transdisciplinary space, not just the primary discipline. Should summarize methodological innovations that are distinct from those of any of the primary disciplines