My workflow with academic literature

When I had to (re-)write the literature review for my dissertation, I restructured my workflow to make my literature management more efficient.1 Throughout the past years, it always felt as if I was accumulating masses of knowledge and tons of papers – and for sure, I did! But it turned out to be so much knowledge that it was hard to keep track. I always liked my way of sorting things (I heavily relied on mind maps to sort the literature) but realized that it was not enough when categorizing, and – most importantly – logically structuring a condensed literature review that presents all the knowledge to my readers in an engaging way.

In short, I had to come up with a system that allows me to track what I have read, what research tells us (and what we don’t know), and then present it in a good structure that makes it easy to follow.2

To tackle this task, I followed these steps:

  1. Save your literature with a systematic name such as AUTHOR_YEAR_SHORT_TITLE.pdf. I know, this is what you’ve probably been told since the beginning of your studies, but it is too easy not to follow the golden rule and instead have tons of papers without a meaningful name stored somewhere in your Downloads folder…
  2. The Downloads folder leads us to the next essential point (and this holds even if you have a good search function on your computer): Save your literature in a meaningful place such as a specific project folder or a watch folder.3
  3. Tag the literature. I use Apple’s built-in tags to find my literature wherever it is on my computer.

smaller_image

  1. Now comes probably the most important part. I am a visual person, and I love mind maps to draw the literature’s general structure. All the time and energy that I invested during the past years in sorting my literature in mind maps was well spent – it was easy to combine them all into one large mind map. But you can already tell from the screenshot that it is not the most accessible system when you have much literature (if I printed it in a readable size, the whole mind map would cover a few square meters – and even in an office building it’s hard to find a wall that you can cover with it!). Mind maps are great for visualizing the relationship between the papers and you can zoom in on the computer and get more details, but they are less sophisticated when you also try to get a good (and more abstract) overview of other characteristics such as theory, methods, or data resources. A tabular overview form might be better – even more so if the tabular overview allows you to filter the data. I used Notion for this purpose and liked how intuitive it is. But it has its limitations when it comes to printing (which is not so straightforward and user-friendly), and I am sure Excel or other tabular processing programs also do a good job.

small_image

In retrospective, I should have started the tabular overview earlier and – at least for me – an ideal program would allow me to combine all steps in one: Store the references; have different categories where I can take notes (such as Research question, Theoretical argument, Dependent variable, Independent variable, Methodology, Findings, …) which also converts it in the tabular overview; and allow me to present and connect the literature as a mind map if needed.

  1. And, as the last step, save your references so that you can use them in your text processing program. I use JabRef to store my TeX citations but I have also tried Mendeley and Zotero in the past. It is easy to sync your references across the programs, so it’s worth trying out multiple of them to find the best for you. I came back to JabRef because I experienced it as most stable and user-friendly (Mendeley kept crashing and lost the saved citations; Zotero does not allow you to copy-paste plain LaTeX code for the citations – a workflow that I use regularly). In JabRef, you can manually add citations or import them using the DOI or LaTeX code from Google Scholar.4 It is not super fancy but more minimalistic and, most important: stable and reliable – and that’s what I like about it.

And that’s it – now let the writing part begin!

small_image

  1. This post contains links to products that I use in my regular work life and bought myself. As always, these workflows are never a one-size-fits-all solution and present just a snapshot of what currently works well for me. ↩︎

  2. If you are interested in a data workflow for small(er) projects, here is a blog post that I wrote on it. ↩︎

  3. I came across the idea of a watch folder when using Mendeley. Although I am no longer using it, I have a watch folder where all documents that still need to be added to my reading program (GoodNotes) are stored. ↩︎

  4. Ever wondered what LaTeX is? Dennis Hammerschmidt and I wrote a short and hands-on blog post on how to get started with Overleaf and LaTeX. ↩︎