Making sense of the stuff you have collected during the initial stages of your PhD thesis

«He had a plan. Maybe you just didn’t see it ’til it hit you between the eyes. But, it started to make sense … in a Tyler sort of way. No fear. No distractions. The ability to let that which does not matter truly slide.»
Narrator in «Fight Club» (1999)

While research is often seen as “I go from point A to point B” (e.g., something does not exist, I make it exist, or we don’t know something, I find out something), in reality, it’s often an anything but linear process (I recommend Alon, 2009, here).

Things are not linear. You might start off with a topic, e.g., to develop a tool or solve a problem, but when working with the topic — reading about it, attending conference, advising bachelor and master theses, or student projects, doing research and writing up your findings — the topic morphs.

In many cases, exploring the topic, esp. when advising students, isn’t bad. On the contrary. It’s only bad if the day to day work prevents you from engaging with your PhD topic. But exploring the area around the topic you’re working on, that’s a boon.

The problem starts, however, when you want to nail down your thesis. You collect lots of material, all somewhat related to your thesis. But what is it you are actually doing? Then the question becomes: How do I make sense of all the stuff I did? Was it all peripherally related to the topic? Or did actually shed some light on it?

I think there are two approaches to find out: Bottom up or Top Down.

Bottom Up: «Hello data my old friend …»

Frankly, I don’t believe in silver bullets, but if there is anything I really believe in, it’s content outlines. Just collect the data you have and sort it bottom up.

For example, you have written a couple of shorter papers and advised a few theses. Great. Now cut them to pieces. Whether it’s by paragraph (if the students did know how to use them) or by sentence, break the works down. Many sentences will stick to each other (e.g., sentences starting with “This means …”, “Thus …”, “Consequently …”), so paragraphs might be the saver bet (you might have to split them down further). Then group the paragraphs/sentences meaningfully (e.g., you have written something about prerequisites in 2 different papers, the information should end up in the same place.

Outliners (e.g., OmniOutliner on the Mac, UV Outliner under Windows) are great for this task. For example, the information would end up as child cells of the same parent cell (prerequisites). Make sure you know where the paragraphs/sentences did come from. For example, you can replace the paragraph mark (^p in Word) with “[authorname_year]^p” to have the name of the author and the year attached to each paragraph. Same if you first split the text into its sentences (replacing “.” and “?” and “!” with “.^p” and the like). In outliners it’s probably best to use a dedicated column for this information. No matter what you use, you always need to know where your sources come from.

Sorting them bottom up isn’t easy, you risk forming the wrong categories or sorting information that belongs together into different places. But it can be done. It’s detail oriented work well suited for Aspies, but it can be tedious for the rest. Still, it is one way to find out what you have, working from the data to build up categories of information related to your research topic.

Top Down: «I never did mind about the little things.»

The other way is to ignore the details for the moment and focus on the big picture. Start with the central topic, e.g., on a mind map. Then go through your (advised) works and identify the main topics. The structure of the texts — mainly headers and sub-headers — will help here. You get information, e.g., about analyses, or about methods, or about different aspects in the theoretical background. You identify the major sections and divide them further and further. For example, which topics are discusses in the introduction and discussion. How can you group them further?

Going top down might be easier for people who don’t like details, but it might also lead to missed details. Of course, those don’t matter that much in this approach. And as written, mind map software might be useful here. You can usually add some text in notes fields.

What both have in common: It will take time.

Making sense of the material you have isn’t an easy task — and it’s not quick either. You need time — need to make time and defend that time. Depending on the amount and quality of the material, it can be done in two to four weekends. Working weekends. It should not take longer if you actually spend the time working. After four to eight distraction free workdays (hint: get something to drink and some non-stick/non-greasy finger food) you should see some patterns.

Of course, after the patterns become visible, you still have to decide what to do next. Are there some areas on which you should focus? Do you have enough for an overview? Perhaps a chapter or two?

But no matter the findings, it’s good to know where you stand. Especially if you are positively surprised.

Alon, U. (2009). How To Choose a Good Scientific Problem. Molecular Cell, 35(6), 726–728.