Anyone who has the misfortune to be near me when I’m doing research or who follows my twitter stream (@pyoor) will probably have encountered my rants about doing literature searches and finding papers.
Essentially, it boils down to this. It doesn’t work very well.
For those of you who’ve not had the joy of searching the literature (by which I mean the miriad journals of medicine) there are several (many) databases, all of which contain a subset of the available journals which you might want to examine.
If you are conducting any kind of proper literature search then you have to search multiple databases to attempt to ensure you’re getting all of the available data on any given subject.
Each database is subtly finicky and requires it’s own version of what you’re searching for to make sure you’re actually searching for what you want, and not say, topics that are distantly related to what you want. Or missing great chunks of what you want because it wants a different wildcard after the search term so you get words including that term.
Then you have to narrow each search individually to, say, human studies, or studies done in the last five years.
Then it gets really fun.
See, some journals have their text available online. Some don’t. For this bit of my course, essentially if it’s not online I can’t see it, because the time it takes to get a paper (minimum of days if it’s not in my work library) in dead tree form means it’ll arrive after I’ve finished trying to answer the question. However if it is online, the excitement just continues.
You might think I’d click ‘full text’ and lo, there it would be.
Ha.
Ha ha.
Ha ha ha.
No.
No, what happens is this:
Click ‘full text’. Oh, not held by my uni’s library.
Go back to database. Click on find full text, follow succession of links. Oh, only available to buy online.
Find journal on site. Oh, that’s a Science Direct, or Ovid, or Whateversville journal. Attempt a further login (“institutional login”) and find that…no, it’s not available.
Check if it’s on the list of journals which the university individually provide access for, which aren’t accessible through the access federation (an organisation which provides access to lots of journals) – it’s usually not because it’s quite a small list.
Pause, search Google. Find that…no, it’s not available.
Log in to Athens, which is what the NHS uses to provide full text access to journals.
Find link to journal.
Find that I don’t have access to this particular journal through the NHS either.
Swear.
Repeat for next article.
Each time it takes me minutes of searching, sometimes, at some point in the process I find success. Sometimes it’s been republished on an author’s site as a handy, free, usable PDF.
At every stage though it’s frustrating. Really, truly, this should be a three step process.
Search. Select. Read.
Search should search all the available databases, and should exclude multiple results for the same article (always hilarious). And should do so from one place.
Select should allow me to see, in one place all the articles I have access to under every login that I have. My uni provide logins for things like ‘The Lancet’ which aren’t available thorough my databases and which aren’t provided thorough the federation. It’s a complete pain in the arse to have to go and get the password and username for these each time.
Read should be just that.
And then I could spend my time swearing about the poor quality of most papers, as opposed to swearing about not being able to find them.
