Taking Reading Notes in the Social Sciences

As in every discipline, social scientists need to read and take reading notes. I keep all of mine in Zettlr. Here I describe how I take them.

Preconsiderations

Firstly, I have created my own workspace in Zettlr where I store all my reading notes. That is: I created a folder inside my Nextcloud (you can also use Dropbox, OneDrive, or just have a folder without syncing it anywhere), and then I pressed Cmd/Ctrl+O in Zettlr to open that folder as a top-level workspace.

This way I have all my reading notes in a single space, and I can limit the global search to that specific workspace to only search in my reading notes.

Creating Snippets

Next, it is important to note that in my area (Analytical Sociology) we have broadly speaking two types of texts to work with:

1. Theoretical or methodological journal articles or books
2. Studies

To save time when taking reading notes and to make sure that they follow a common structure, I have created two templates in Zettlr.

For this, I went into the assets manager and created two snippets, called reading-note and review-template. Ignore the naming, the names don't really reflect what I use them for, but since my brain has already associated the names with their contents, it is easier to just stick to them (because I'm the only one who needs to distinguish the two).

The reading note template is very plain and does not add too much structure. It looks like this:

---
title: "${1:Author (Year) — Paper Title}"
date: $CURRENT_YEAR-$CURRENT_MONTH-$CURRENT_DATE
id: $ZKN_ID
author: Hendrik Erz
---

# ${1:Author (Year) — Paper Title}

$0

## References

- ${2:Citekey}

As you can see, it already includes a frontmatter, it auto-generates a date and, just in case, it contains an ID.

Note as well that there are two tabstops with number1. The reason is that for metadata reasons it makes sense to have a YAML frontmatter title, but within Zettlr it is easier to see the first heading level 1, so I literally duplicate the contents here. An added benefit is that no matter which way I choose to display my files in the file manager (filename, h1 or title), the reading notes will stay the same.

I have added the fields in a useful manner, because I want to immediately add the text's title and authors, before also providing the citekey. And afterwards, I want to get into actually taking the reading notes, hence is the $0 (the final tabstop) between title and references section.

---

The second snippet, review-template, is a bit more complex and prescribes more structure:

---
title: "${1:Author (Year) — Paper Title}"
date: $CURRENT_YEAR-$CURRENT_MONTH-$CURRENT_DATE
id: $ZKN_ID
author: Hendrik Erz
---

# ${1:Author (Year) — Paper Title}

$0

## What Does the Author Want to Know?

1. **Originating Question**: $3
2. **Central Question**: $4
3. **Specifying Questions**:
    - $5

## Why? Or, So What? / Intellectual Rationale

${6:Why does it matter?}

${7:Audience}

${8:Broader Implications of the paper?}

## What's the Author's Answer or 'Argument'?

${9:General underlying theory, or general argument}

${10:Answer to the central question}

${11:Answer(s) to the specifying question(s)}

## How does the Author go about Finding the Answer?

${12:Were any crucial simplifying assumptions made?}

${13:Evidence presented, that is: The data they use}

${14:What measures do they use?}

${15:Justifications for the methods they specifically used?}

## How does the Author Probe or Check whether the answer is right or not?

${16:Measure of reliability, i.e. did they validate the data against itself by computing different measures, hoping they exhibit the same patterns?}

${17:Also, did they check with competing theories if their results are plausible?}

# Reference

- ${2:Citekey}

As you can see, the frontmatter is the same for both. Note that I have added a final tabstop ($0) between title and first section, so that, after I have finished creating this reading note, I can add some final/general thoughts at the top of the document which I can parse quickly if I come back to this note in the future.

Creating a reading note

Whenever I read something, I create a new reading note for that. I do that either while reading, or after reading. To do so, I right-click the reading notes workspace, select "New file" and use as its name the format I also use in the titles: Author (Year) - Short title.md. I trim the title because I want to reserve the ability to open that file on Windows as well, and Windows doesn't like too long filenames. Then I press Enter to confirm the filename.

After the file has been created, I immediately type :reading-note or :review-template to fill in the snippet and to be able to Tab my way through the different structural elements.

That is where I have to differentiate beetween the two types of reading notes I take:

For theoretical/methodological papers and books

For theoretical or methodological papers as well as for books I use the more open :reading-note template, because what I will be writing down for that is not clear in the beginning. So after typing in the Author (Year) — Title trinity and providing the (Zotero) citekey in the References section, I basically just write whatever I feel is important. I am constantly trying to trim down the amount of stuff I note down, but that's kind of an ability that you learn like riding a bike. Reading notes from my undergrad were horrible and I even split them up into folders because I was writing so much, but by now it's all single files with just the gist of the argument.

What I make sure, however, is to always include the citekey as a proper citation behind everything I note in the form [@Citekey, p. 12345] because this way, whenever I need to copy that into a longer text, I have the citation already as I need it. Saves a few seconds of time later down the road.

For Studies

This now leaves the study papers. You will see that that reading note template is incredibly structured. I have taken the structure from my supervisor who gave it to us in a course he taught. He himself has taken it from Maurice Zeitlin’s The Four Questions. What I did was split them up and mold them into the template.

The aim of that effort was that it forces me to not write too much, because those studies are most of the time solely applications of said methodological and theoretical papers, so what I'm interested more in with these studies is what kind of data they used, how they used it, and what their results were.

So what I will do is first read the study, and only afterwards create the reading note and tab through all of these questions, answering them one by one, if applicable with some sources.

The cool thing about Zettlr's snippets in this regard is that I have the actual questions in the template, so the questions are visible, but when I tab into one of them, they get selected and therefore overwritten. The reason is that I don't need to remember the questions afterwards since my answers are more important, but this way I make sure that I never forget something.