Search strategy

flickr photo by malik ml williams http://flickr.com/photos/brothaloveimages/370413613 shared under a Creative Commons (BY-NC) license

Yesterday’s workshop, ‘Literature searching for researchers’ delivered by members of our library team provided the kick in the pants I needed to fettle my approach to literature searching. Until now, I’ve never had any trouble finding literature; rather it’s more a case of finding ways to manage the vast quantities which have emerged. What became clear to me during the session however was that I need to develop a more systematic strategy; one which is planned, clearly articulated and open to scrutiny. I haven’t been worried that I’ll not have covered sufficient ground to bring the significant themes and important issues to the surface with my ad hoc, piecemeal literature searching. However, at this level, I need to be sure that there’s not something I might have missed which might come back to bite me during and future presentations I might give, or worse still, during a viva.

This post will begin the process of thinking about my strategy and documenting how that unfolds so I have a reference point to which I can return. This should be framed quite naturally by my area of research and the questions I want to answer. The main themes centre on teacher professional learning and Twitter, but this will be undertaken using an actor-network theory approach. In a loose way I’ve already begun that process when moments have arisen; if a particular article or reference within it caught my attention, I’d follow that up through our Library Gateway or using Google Scholar, then the search process would capture my attention for a while and I’d spend a brief period extending the parameters. However, I never documented those forays, so found myself repeating them, either intentionally in order to explore the areas more thoroughly, or unintentionally because I simply forgot that I’d done it previously. One thing the workshop helped with was how you might combine and adjust your search terms to return more meaningful results, achieve a greater breadth, or target a particular area more precisely. Importantly, you should then record your activity, both for your own needs, but also since it is sometimes a requirement that you provide details of your search strategy in work that you publish.

I’m no stranger to using boolean search terms, but again, being more strategic and thoughtful about combinations of search terms is likely to yield more reliable and valid results. Having established my starting point, I then need to establish the most appropriate databases and finally devote a good chunk of time to conducting the search thoroughly. By knowing in advance what the parameters of my search will be, I can ensure that the searches I conduct in different platforms will be consistent and that if I need to retrace my steps later, I can do so. Importantly this also speaks to some on the principles of research integrity; laying your research techniques open to scrutiny, but also making the steps you took clear to anyone who may wish to replicate or extend your research in the future.

Our Library Gateway search facility casts its net widely amongst other databases; a sort of meta-search if you will. This will provide the more general point of departure for the search expedition, but be followed up by more narrowly focused locations to hone in on the specifics. This means working down through multidisciplinary databases like Web of Science, Directory of Published Papers and Google Scholar; through databases more closely associated with my field of interest like ERIC, the British Education Index or filtered versions of targeted databases like ScienceDirect; then into specific journals like The Journal of Educational Research or the International Journal of Actor-Network Theory. Although often unpublished, another particularly useful source I’m keen not to neglect are theses and dissertations. I can search for these through EThOS or Proquest, not least because the final product of other peer’s labours will also help provide a sense of the standard and quality to which I ought to aspire.

Having now established where I’ll be roaming, I now need to consider for what I’ll be searching. My springboard will be the three themes of professional learning, Twitter and actor-network theory, so the search terms I’ll start with are:

• “professional learning” OR “professional development” OR “CPD”
• “twitter” OR “social media” OR “microblog*”
• “actor network theory” OR “sociomaterial*” OR “ANT”

(I’ve not tried this yet, but since actor-network theory is commonly known by its acronym ANT, I need to allow for that, but am not sure whether I might end up with a lot of results discussing the Formicidae family!)

I’ll begin by combining all three sections with AND, then take each of the three pairs separately and finally each one individually. Undoubtedly as the searching and research unfolds, It’s likely that those terms will need adjusting, so where allowed, I’ll save the searches so I can simply update them in the future, rather than having to recreate them. Additionally, given that research is an ongoing process, and to avoid missing out on work published subsequent to my search, I’ll take advantage of the alerts feature that most of the aforementioned databases provide. This will email me details of newly produced work associated with those searches. An alternative that some databases provide is an RSS feed for your search; RSS is a technology I’ve been using fruitfully to keep informed for a long while now. Since I already have an RSS reader set up to bring me content from a variety of different areas, I could incorporate my research searching into that.

One strategy I’ve continually made use of is the snowballing technique where having found a particularly significant paper, you follow up the pertinent references contained within, then from those new sources do the same until you’re not unearthing any new materials. Another useful tip I learned in the workshop is citation searching. This involves deploying your usual search terms, but sorting the results (where the database allows) on the number of times cited. Hopefully the significant papers bubble to the top, then the ‘Cited by’ function can be used to spread out to related or connected research.

The final step is to document the research you have undertaken. This if for your own benefit to ensure you are not duplicating your efforts and retrace your steps; for your supervisors so they can keep track of your progress (and be able to highlight any areas for further development?); and for others who may wish to benefit from, or build on your research. The University of Leeds has produced some helpful guidance, including templates and examples of documents you could use.

So that’s it. Plan begun. Now for the execution …