How to stop making more work for yourself – keep track of search terms

You cannot search “The Literature”. “The Literature” is a monster, growing, sprawling and changing at a faster rate every single day. I find myself attempting to search “The Literature” sometimes, when I can’t be bothered to do any real work, but don’t want to feel guilty about not doing anything, I’ll go to a search engine, type in some terms and sift through the thousands and thousands of titles, maybe hoping to glean nugget of gold. Obviously, this never happens. The fact is that even if a golden nugget of an article was present in a search done like this, I would probably overlook it, being drowned out in the overwhelming background of junk.


Doing literature searches in this way is futile, boring, time consuming and time wasting – don’t do it!


A far better strategy is to search for keywords that return a very high proportion of relevant results and then………. record those keywords somewhere safe and easy to find, so you can do the process all over again.

Let me explain. Don’t just type in one keyword, type in two, three or even four to really narrow down the search result to the field you’re looking for. Also, use Boolean operators such as AND, NOT and OR to specify exactly which phrases you want in your search results and which ones you don’t.


Doing this more often than not generates a manageable list of articles for conducting at least a section of your literature review. This is essential. You have to be able to flick through, and make a judgement on whether the article is useful or not, for every single search result. If not, you cannot be sure that you haven’t missed something useful further down the list. If you have used all of the keywords and Boolean operators you can, and the list is still too long, see if you can refine the search (possible in search engines like; google scholar, scopus, web of knowledge, pubmed) by any number of parameters such as year, journal authors etc.

But this excludes loads of articles I hear you cry…. probably some of them are really important to my literature review. And I would agree, but…..
and here comes the really powerful part, once you have the list, record the search term, database used and the refinements made, along with the date you conducted the search in a word file or spreadsheet, related to the literature review that you are doing. And then review, every single article in the search and either mark it/save it for further inspection, or disregard it as irrelevant. By looking at every item in the list, you can now say that you have fully assed the keywords that you recorded and you can go on try other combinations to do the same thing.


But what about the ones you miss????


Well, by doing a literature search in an arbitrary way you will miss far more articles and which is more, you will have no idea why or how you missed them. However, by recording the keywords and doing narrower searches, you can review the keywords that you have searched for in the past and refine future searches to try and change the types of articles you find in the future, knowing that every time, you have exhaustively checked that keyword combination, which represents a small, but very complete portion of the literature. In this way you attack a massive body of literature, small section at a time. The best part is that every time you do a search you add to the searches that you have already done, and never go over old ground. It can take as little or as much time as you like as you decide how many search results you want to review!


Go on give it a try. Set up a document to record your keyword searches for literature review project!