Here’s something I’ve meant to share for a while now. I use Zotero to manage things I read (articles, books, conference papers, etc.), but what follows is applicable to any similar software. I keep notes on everything I read, and over the years this has evolved into something quite structured (a template in fact). The fact that it is structured is quite useful when I come back to a paper after a while. Here’s the template:
Obviously, not all papers will have something for each heading. While the heading research question is rather innocuous, unfortunately it’s not always as easy to fill in as it should be. The dependent variable is the quantity of interest; under explanatory variable I include the main explanations. I tend to include control variables here, too, although in brackets.
Data describes the data sources, such as the survey used, the countries covered, population covered, N; experts, ABM, or even “data free”, whatever seems the most adequate description. Method is for methodological details. While usually we are more interested in the results rather than how they were obtained, a quick glance at the methods (and data) can be really helpful in determining how much weight I want to give a particular result.
The heading mechanism is often challenging to fill in, simply because many papers do not state them explicitly, or because the theory section is not tightly connected with the empirical part. I’m not lamenting here; I guess I’m guilty of this, too…
Often my interest is in the results section, where I summarize the main findings. The heading notes takes everything else, namely free notes.
The whole things is (deliberately) rather flexible, but it helps with two things: (1) read papers with some focus, (2) have notes in a format that allow me to retrieve relevant information more quickly (here the advantage of a database over Anki, but obviously only when things can be found).