Friday 6 November 2009

The Mechanics of Doing a Literature Review

The question of how to conduct a literature review comes up a lot. Things have changed since I spent time in an actual library, reading hard copy articles taken from dusty shelves and photocopying them. The basic principles however, remain the same. It is just that the technology makes it so much easier. First, if you're not a registered student somewhere then Google Scholar is free for the most part and a useful source. If however, you can access library facilities then on-line databases such as Business Source Premier or ABI/Inform are fantastic and tend to be a better way of searching the top journals. Here's a step by step guide ...

  1. identify what you want to look for ... at least in broad terms your research needs an initial focus. Of course reading helps sharpen this focus but you might start with some basic keywords, terms or authors.
  2. use an electronic database to allow you to search for everything that has come up using these search terms in the top ranked journal(s) in your field. For me I always start by looking in AMJ, AMR, ASQ, SMJ, Org Science, Org Studies, JMS, HR, Organization and BJM. These are by most accounts the top general management journals in the US and Europe (except SMJ which is a strategy journal).
  3. Having scanned ... look for who is writing ... identify key scholars then look for what they have written outside the top journal outlets. This broadens your search.
  4. For each key paper that you find look back and forward in time. That is, see who the author(s) are citing as influential thinkers to help you work back to key sources and theories. Second, using the technology, see where your key paper has been cited since it was written. Most of the databases have both a "references cited" and a "cited in this database" tab to allow you to do these two tasks.
  5. For each paper, think about future research areas. Most papers close out with an "areas for further research" which is a good starting point. Item 3 above might help you find whether the author concerned ever followed up on their research.
  6. Next, for each paper, take note of three things. First, what assumptions does the research make. Second, what root theories do they draw upon and contribute to. Third, what methodological stance are they adopting in the research. You should be able to comment on and map each of these three things before moving on from the paper.
  7. Finally, make notes about who said what in which paper. Be thorough and organized. If you're just starting out use something like Reference Manager or EndNote ... it'll save you major grief in 3 years time when you come to try and track down the beautiful quote you want to use without having to re-read every paper you've ever glanced at. Full details ... at the time of writing ... you'll only have to go back and do it again if you don't do it at the time.
For more details look at Chapter 3 of Research Methods for Business and Management.


  1. Well I wish you had mentioned point 6 a little earlier Robert - but seriously good idea to help focus the reading. You talk sometimes about density of referencing - what is your rule of them there?

  2. Its not a rule ... more a feeling. There is something about the way in which good academic papers are written that give a sense that the author(s) really do know their stuff and that they really have read and understood the material that they are citing. In my experience, this comes across as a certain density and perhaps tone of voice in the way that citations are used (see also the Writing Skills post elsewhere on the site).

    Here are two links to things which I'd rate highly. First is an outstanding book by Steve Floyd and Bill Wooldridge reviewing process views of strategy making. You can glimpse enough of its contents on the Amazon link below to see what I mean about the sense of reading something that is well grounded in the literature.

    The second is just one example from an academic journal. I've chosen Kathleen Eisenhardt's "building theory from case study research" in AMR 14(4), 1989 ... for two reasons. First it is probably one of THE most heavily cited case study articles and anyone thinking of writing in cases should read it (along with the Case Study post elsewhere on this site of course!). Second, she writes clearly and articulately. Notice that in the AMR article, the citation density is a little higher than in the Floyd and Wooldridge book. It is probably just a result of the word count restriction of a journal format. You can access the full paper via on-line resources at your university but the front page of the article is available at the link below.

    Hope that this helps ... Robert

  3. Great informative post you've shared with us, and I wanted to have an opportunity to ask if you could take a quick look at the role of literature review here, and if it's possible please provide your valuable feedback on this.

  4. Dear Divyansh ... thanks for your comment and for the link. The other posting offers a few specific sub-types of literature review (argumentative, integrative, historical, methodological, systematic and theoretical) so readers might find it useful to establish which kinds of review are the norm in their discipline. This could be done by reading some recent the literature reviews of some recent PhD completions or by looking at the peer-reviewed outputs in the journals that you read and cite. The other observation is that the other post seems a little more appropriate for those areas of research where definitions, methods, contributions are more easily or consistently defined. The advice I set out in my own post offers my variation on a systematic literature review.

    Good luck with your own research