Swath and Dive: A pattern for PhD defense presentationsby Luis P. Prieto, - 10 minutes read - 2037 words
In recent times I’m having the fortune of seeing several of my own doctoral students approach the end of the doctoral journey (yes, it does end!). As they submit the dissertation and prepare for their defense, there is one piece of advice I find myself giving again and again, about how to tackle the impossible task of presenting multiple years of research work in less than one hour. In this post, I describe a “presentation design pattern” for thesis defenses, which builds upon classic conceptualization exercises advocated in the blog. I also illustrate it with an example from my own thesis defense presentation, more than ten years ago (gasp!).
I still vividly remember when I had to prepare my defense presentation, how I tried to shoehorn tons of concepts into an impossibly small number of slides… which still were too many for the 45-minute talk I was supposed to give at the defense. After several rehearsals (with an audience!) and lots of feedback from my colleagues and advisors, I finally stumbled upon a solution. Later on, I have found that a similar structure was also helpful to other doctoral students preparing their defenses.
The rest of the post takes the form of a presentation design pattern, i.e., a description of “a problem that occurs over and over again in our environment, and […] the core of the solution to that problem, in such a way that you can use this solution a million times over, without ever doing it the same way twice."1 (a concept originally proposed in architecture, and later used in software engineering, pedagogy and many other fields). I have called this pattern Swath and Dive (for reasons that will become obvious in a minute).
The context: when is this pattern applicable?
When you have to prepare an oral presentation for a doctoral dissertation defense. This pattern is especially helpful if the research is a bit complicated (e.g., composed of multiple contributions, multiple studies, or using multiple research methods) and it is not obvious what contents to include/exclude from the presentation.
What is the problem? What forces are at play?
The main problem this pattern tries to solve is the seeming impossibility of showing 3+ years of research work in less than one hour. While time restrictions and structure for the defense are different in different countries, typically 25-60 minutes are allocated for the presentation. This limited time is a key force at play, but there are others as well:
- The sheer volume of a thesis dissertation’s contents (typically, a 100-500 pages document), which itself is a condensation of years of hard research work.
- Defending PhD students need to prove to the jury that they are now competent, independent researchers (i.e., they master the literature of their topic, are able to apply a research methodology and think critically about the results ).
- The varying levels of expertise and familiarity of the jury members with the concrete thesis topic.
- The varying levels of knowledge that jury members have of the dissertation materials (i.e., did they read the dissertation document in full? with what level of attention?). While all members are supposed to have read the document, in practice there is a lot of heterogeneity in compliance.
The typical end product of these forces is what I call the “skimming” approach to the defense presentation (see picture below): The presentation provides only a very high level overview of the main elements of the dissertation document (sort of like a table of contents). More often than not, too much time is spent in the introductory and related literature parts of the presentation (which are somehow “safe”, less likely to be criticized – another instance of avoidance at work in the PhD), and time runs out when the student is getting to the really interesting part for the jury (the student’s own work). This approach of course has the critical flaw of not showcasing enough of the student’s own abilities and research outcomes.
How to avoid “skimming” your dissertation? Enter Swath and Dive.
The solution: Swath and Dive
What I propose in this pattern is to structure the presentation in a different way, a way that tries to balance the need for an overview of the dissertation and (at least some of) the richness of the investigation and the hard work the student has put behind it. The proposed structure goes like this:
A swath is “a long broad strip or belt” of grass, often left by a scythe or a lawnmower. In the context of a dissertation defense presentation, this is where the student gives the overview of the main elements of the thesis: key related scientific literature, main research questions, contributions to knowledge the dissertation makes, etc. Long-time readers of the blog will recognize these key elements as the components of the CQOCE diagram, one of the key reflection exercises in the “Happy PhD Toolkit” to (iteratively) understand and discuss with supervisors the overall view of the thesis. Aside from those key elements, probably some notes about the research methodology followed (which are not part of the canonical CQOCE diagram exercise) will also be needed.
In a sense, the Swath is not so different from the typical “skimming” mentioned above. There are several crucial differences, however: 1) when developing the Swath, we need to keep in mind that this is only a part (say, 50%) of the presentation time/length/slides; 2) the Swath should give equal importance to all its key elements (e.g., avoiding too much time on the literature context of the thesis, and making the necessary time for the student’s own research questions, contributions and studies); and 3) the Swath does not need to follow the chapter structure of the dissertation manuscript, rather focusing on the aforementioned key elements (although scattering pointers to the relevant chapters will help orient the jury members who read the dissertation).
Then, within this high-level Swath describing the dissertation, when we mention a particular contribution or study, it is time to do…
This part of the presentation is where the student selects one study or finding of the thesis and zooms in to describe the nitty-gritty details of the evidence the student gathered and analyzed (if it is empirical research), how that was done, and what findings came out of such analysis. The goal here is to help the audience trace at least one of those high-level, abstract elements, all the way down to (some) particular pieces of the raw data, the evidence used to form them.
How to select which part to Dive into? That is a bit up to the student and the particular dissertation. The student can select the main contribution of the dissertation, the most surprising finding, the largest or most impressive study within the work, or the coolest, most novel, or most difficult research method that was used during the dissertation process (e.g., to showcase how skillfully and systematically it was used). The student should give all the steps of the logic leading from low-level evidence to high-level elements – or as much as possible within the time constraints of the presentation (say, 30% of the total length/time/slides).
An essential coda: Limitations and Future Work
Although this didn’t make it to the title of the pattern, I believe it is crucially important to keep in mind another element in any good defense presentation: the limitations of the student’s research work, and the new avenues for research that the dissertation opens. These two areas are often neglected in crafting the defense presentation, maybe with a single slide just copy-pasting a few ideas from the dissertation manuscript (which were themselves hastily written when the student was exhausted and rushing to finish the whole thing). Yet, if the student convinced the jury of her basic research competence and knowledge during the Swath and Dive part, a big part of the jury questions and discussion will focus on these apparently trivial sections.
When doing the limitations, the student should gloss over the obvious (e.g., sample could have been bigger, there are questions about the generalizability of results) and think a bit deeper about alternative explanations that cannot be entirely ruled out, debatable aspects of the methodology followed… squeeze your brain (and ask your supervisors/colleagues) to brainstorm as many ideas as possible, and select the most juicy ones. For future work, also go beyond the obvious and think big: if someone gave you one million dollars (or 10 million!), what cool new studies could continue the path you opened? what new methods could be applied? what experts would you bring from other disciplines to understand the phenomenon from a different perspective? what other phenomena could be studied in the same way as you did this one? Try to close the presentation with a vision of the brighter future that this research might unleash upon the world.
To understand how this pattern could look like, I can point you to my own thesis defense presentation, which is still available online. This is not because the presentation is perfect in any way, or even a good example (viewing it today I find it overcomplicated, and people complained of motion sickness due to its fast pace and Prezi’s presentation metaphor of moving along an infinite canvas)… but at least it will give you a concrete idea of what I described in abstract terms above.
If you play the presentation, you will notice that the first few slides (frames 1-6) just lay out the main construct the dissertation focuses on (“orchestration”), the structure of the presentation and its mapping to dissertation chapters. Then, the bulk of the presentation (frames 7-117) goes over the main elements of the dissertation according to the CQOCE diagram, i.e., the Swath part of the pattern. Within this high-level view of the dissertation, I inserted a short detour on the research methodology followed (frames 25-28) and, more importantly, several Dives into specific findings and the evidence behind them (frames 43-48, 66-72, and 99-112). Then, frames 118-136 provide the conclusive coda that includes the future work (but not the limitations, which were peppered through the Swath part of the presentation – a dubious choice, if you ask me today).
Variations and related patterns
As you can see from the example above, one does not need to follow the canonical version of Swath and Dive (mine is rather Swath and Three Dives). Yet, paraphrasing Alexander, that is the point of the pattern: to have the core of the idea, which you can use to produce a million different solutions, tailored to your particular context and subject matter.
It is also important to realize that this structuring pattern for thesis defense presentations does not invalidate (rather, complements) other advice on preparing scientific presentations2,3,4 and thesis defenses more specifically5. It is all very sound advice! For instance, once you have the structure of your Swath and Dive defense presentation, you could use the NABC technique to ensure that the Need, Approach, Benefits and Competition of each of your knowledge contributions are adequately emphasized. And you can rehearse intensively, and with an audience able to come up with nasty questions. And so on…
May you defend your thesis broadly and deeply!
Do you know other defense presentation structures that work really well in your discipline? Have you used Swath and Dive in your own defense successfully? Let us know (and share your examples) in the comments area below! (or leave a voice message)
Header image by DALL-E
Alexander, C., Ishikawa, S., & Silverstein, M. (1977). A Pattern Language: Towns, Buildings, Construction (Vol. 2). Oxford University Press. ↩︎
Carter, M. (2013). Designing science presentations: A visual guide to figures, papers, slides, posters, and more (First edition). Elsevier/Academic Press. ↩︎
Anholt, R. R. H. (2009). Dazzle ’Em with Style: The Art of Oral Scientific Presentation (2nd ed). Elsevier, Ebsco Publishing [distributor]. ↩︎
Alley, M. (2013). The craft of scientific presentations: Critical steps to succeed and critical errors to avoid (Second edition). Springer. ↩︎
Davis, M., Davis, K. J., & Dunagan, M. M. (2012). Scientific papers and presentations (Third edition). Elsevier/Academic Press. ↩︎