% Tools for Academic Literature Review and Analysis
Keeping track, and making sense of, one or more academic literatures is a difficult task greatly helped up electronic tools: most notably research databases and bibliographic managers. I present here how to use both of them here, with a particular focus on my tool of choice: Zotero. Feel free to use any tool you like, but I choose to demonstrate Zotero because it:
- has all the features we need
- allows your data to be synced to the cloud, and so available on different computers
- is completely free; no cost and open source
- is cross-platform and application, and so works equally well whether you use Windows, Mac OS, or Linux for an operating system, or OpenOffice.org or Microsoft Word for your word-processing.
We have already reviewed that the task of analyzing a little consists of the following steps:
- identifying relevant keywords
- gathering references
- reading sources and categorizing them
- describing the connections among the different groupings
We will thus operationalize these steps using Zotero.
Here we first have to figure out what it is we are trying to describe.
Here we will use two journal databases, and Google Scholar, to find as many relevant references as we can.
Once we find these references, we want to quickly suck the relevant metadata into Zotero for later reuse. we thus simply click the button in the toolbar, and Zotero will save the information.
Once we are done collecting the references we can make a quick first pass by tagging each item consistently based on what we can tell from the abstract and a quick glance at the article.
Now that we have everything properly tagged, we want to identify key readings for further analysis. We might want to prioritize those readings that have two characteristics:
- roughly 3-7 years old
- relatively high citation counts (greater than 20 or so in Google Scholar is good; more than 50 is ideal)
More recent articles (or books) have often not been out long enough to become established in the literature, but older articles by definition exclude more recent work. So this is a good sweet spot.
You can then work backward from there, observing how each article builds on previous work. As you read more, you will start to recognize pattens of citation: this article gets cited when an author wants to make X point, and another article when they want to make Y point. Recognizing these patterns is essential to understanding a literature.
In the end, we want a handful of collections that describe different areas of our literature. We want each collection to have roughly ten examples, and for each of them to have a clear statement about why they are there, and what their significance is. At this point, we can ask Zotero to generate a report for us.
Once we have this, we can start work on our papers. Each collection should ideally map onto a section heading, and each article should get cited in that section.