Conducting a Literature Search.

Introduction.

A literature search is a standard component of any piece of research. Once you have established a basic area, and a defined problem, which you wish to investigate, it becomes necessary to know - in exhaustive detail - what previous work has been done in the same area. There are several reasons for this. Partly it is a matter of educating yourself in your chosen area of research, partly of ensuring that you are not duplicating work which has been done elsewhere, and partly an attempt to build your work on sound foundations which others have laid down.

This rather idealistic picture is complicated by several factors. Firstly, it has to be borne in mind that you are writing an undergraduate dissertation - not a Master's or Doctoral dissertation - and the standard requirements for a literature search have to be relaxed. Secondly, recent decades have seen an explosive growth in 'knowledge', which is now the standardised product of an academic industry running completely out of control; there is more literature out there than you could ever hope to survey (even in a narrow field), and a fair percentage of it doesn't deserve to be searched for (I know, because I have written some of it myself). Thirdly, as an institution we have very limited library facilties; whilst you can search on-line databases for articles and order them, it is unfair to expect poor students to spend £1 per document on information they will probably never need.

These notes must be read in the light of these warnings.

The Classic Literature Search.

The classic literature search is likely to be carried out in the early stages of some form of funded research: a Ph.D. or some sponsored research. Those carrying out the research are likely to be thoroughly grounded in their specialist area, and they will have a clear research question which they are pursuing. The next stage is to spend a large amount of time (possibly a year of a Ph.D.) with a comprehensive paper or on-line catalogue and abstract source, firing off keywords relevant to the research area; and then following up the most promising leads in a copyright library. The end result is, in theory, a comprehensive picture of what everyone of any note has ever said on that specialist subject.

In this process it is assumed that much of the information will be coming from learned journals specialising in the subject being researched, or from conference proceedings. At the same time, our classic researcher is likely to be attending conferences (and now partaking in email discussions) on the specialist subject. Our researcher will be getting to know key figures in the field and becoming part of a research community. All of these factors will help to build the 'mental map' of significant research which will be surveyed and quoted in the final research report.

It need hardly be said that this is rather more than we expect of you.

A Realistic Picture of Literature Search.

The following, three-pronged strategy is likely to yield as good a result as I have seen in an undergraduate dissertation. Carry out each step thoroughly; keep notes and a diary. Learn how to keep accurate book references.

Text Books and Course Notes.

Probably the best place to start a limited, undergraduate literature search is an up-to-date text book (published in the last couple of years). In any field which you have chosen, there is likely to be such a book (or books). These books can do two things for you: they can educate you in your chosen subject, and they can lead you to most of the major works which have shaped your field. Study the bibliography(ies); see how much of the material we have in the library. Make notes on (or photocopy - the expensive option) as much of the material as you practically can. If our library is rather thin in your chosen subject, visit a better library or spend a day in a London bookshop.

I take it for granted that you will use our library catalogue to search for suitable material.

Standard Journals.

Our library is not completely devoid of journals. We try and keep the major journals in each subject. Going through back issues and collecting relevant articles can produce a useful overview of a subject and provide a bibliography for further search.

Internet.

Given the uneven quality of much of the material on the Internet, and also given that sheer volume is likely to drown you in information, I would suggest that you use the Internet after you have had an initial attempt at the previous two sources. You will be better placed to sift information when you have been through the previous two exercises. Try and stick to information from e-journals and conferences, or work from recognised academic departments (search mostly .edu and .ac urls).

Literature Search as Cognitive Model Building.

The point of a literature search is not simply to shuffle paper. You are trying to build a mental model of a field and the work done by the key figures in that field. It will probably help to try and build an explicit paper map of your chosen field. Something like a 'mind map', adapted to carry key names and standard references would be a valuable tool. A high quality 'existing theory' section of a dissertation can be written directly from such a map.

Referencing.

While you are searching for relevant literature, try and absorb the various referencing standards that are commonly used. You will be required to reference material accurately in your dissertation. I suggest that you use the Harvard referencing approach - a leaflet should be available from our library.