Reading #3--accessing the literature

If you read "Reading #2--reading the literature ("The Literature"), you probably were wailing something along the lines of "But I just did a Google Scholar search in my own field and there are thousands of papers!  I don't have time to read all those!".  You're right.  You don't.

First, the good news:  Most of those papers will not be relevant.  

Now, the bad news:  You need to at least scan then to determine their relevancy.

Now, the really bad news:  Such searches can overlook papers that are more important and relevant, and not just because your search terms weren't adequate.

It's enough to make one throw in the towel.  Don't.

Before I start on the topic of this blog entry, let me clarify what I mean by "accessing".  I'm not talking about the technical side.  I'm assuming you know that part.  If you don't, go to your library and make them teach you.  However, I can't avoid the old trope "when I was a kid....". You've got it easy, so quit complaining.  Back in the day, we relied heavily on gigantic volumes at least 8" x 11" x 5", with pages as thin as tissue, printed in 6-pt type called Science Citation Index to search the literature.  I would regularly end up with back strain after searching the literature with those puppies (OK, the part about the back strain is an exaggeration).  The idea of sitting at a computer with all that information at our fingertips was still the stuff of science fiction.  Of course, I will concede one way we actually had it easier:  The literature was nowhere near as vast as it is today.

So how do you tackle all this literature?  It takes practice.  As you advance in your career, you will find it easier to just read through the list and know which articles are very unlikely to be helpful.  But in the meantime, here are a few tricks:

(1)  Start with a few papers you know are relevant, the more recent the better.  Read those papers with an eye towards what papers they cite.  You can almost always tell from the context of the citation whether the cited paper might be useful.  Go read those papers.  And, in the ones that turn out to be relevant, pay attention to which papers they cite.  And so on.  Yes, this is time consuming.  All good scholarship is.

(2)  The first time was working backward through time.  You should also work forward.  If you know of an older paper that is significant, find out who has cited that paper.  "Web of Science", which is available through many if not most academic libraries (and basically replaced the old Science Citation Index) is very helpful for this.  Look up the old paper in "Web of Science" to see who has cited it.  That list is very likely to contain at least a few more-recent papers that will be relevant to your research.

(3)  When you find a paper you think will be relevant, read the abstract and skim the rest.  That will usually tell you if it's worth reading more deeply.  If you can't figure out why it was cited and suspect you're missing something, read it more closely with an eye toward which part is important to you.  If it was cited in connection with the methods, read the methods; if the data, look at the data; if the conclusion, read the discussion.  If you find the paper seems to be quite relevant, read it through.  Sometimes first appearances can be deceiving.  I will discuss the perils of citation in another blog post.

(4)  Every once in awhile, read something that seems a big off topic.  You might be surprised what you might learn, and that may open up whole new lines of research for you.

These tips will help you sort through the literature, but you still much remember what I said at the end of point #1:  All good scholarship is time consuming.  Bear in mind that you must really read many of the papers you find.  And think about them.  And think about how they bear on your work.  There are no shortcuts for this part of scholarship.  If you try to take shortcuts, it will show.