Thursday, 15 May 2014

Pointers on writing a #literature review from M. Hammersley

Seminar on reviewing literature in social and educational research given by Martyn Hammersley, CREET, the Open University.

Although literature is seen as an important part of the research process, there are some criticisms from non-academics on its transparency. Martyn provides an overview of practices one can consider as a researcher to make literature review chapters more accessible, both to researchers, policy makers and interested parties.

Systematic reviews and some of their challenges
What policy makers would like to see is a systematic overview, a synthesizing the studies, so including information on how much literature is covered, and whether specific consistent criteria have been adopted

Disciplinary fields and boundaries are much weaker now, then they were before. It is important to focus on literature that is relevant for your own research, and not necessarily limit yourself to what is connected to your particular field. But when you are going to write articles about your research, you need to focus on a specific field to be published in those journals with that particular focus.

Qualitative synthesis
Do systemic review and qualitative synthesis produce literature reviews, or are they forms of secondary analysis done by other researchers (do direct reading).

Another option in qualitative literature is the interpretive critique
  • Traditional reviews attacked for being in conflict with the basic presuppositions of qualitative inquiry, because they:
  • Assume a linear model of the cumulative development of knowledge, in which each study adds a new ‘brick in the wall’
  • Assume that research studies can be objectively assessed in terms of thie methodological adequacy and/or their representational varacity, and assume that these are the most important criteria of assessment (Hammersley 2013, ch.10) 
Key question to ask: what is my literature doing, what needs to be in there in order to prepare for what is to follow in the dissertation (main audience, define boundaries, making judgments about what is more or less relevant).
  • How are you going to go about finding the relevant literature
  • How to evaluate those studies?
  • How are the studies to be selected, and which are to be included in the review
  • How are the boundaries of what would be relevant studies to be defined
  • How is the review to be structured?
(Hart, 1998)

Functions the existing literature can serve in research:
  • It constitutes the context for the formulation and development of research questions
  • It will usually offer possible answers to those questions, and indicate what might count as adequate answers
  • It may suggest ideas and methods that culd be imployed
  • It will provide resources that can be relied upon in developing arguments and evidence in support of answers to research questions
While writing the literature you will have to use your own argumentation, and research goal, making use of the existing literature. But reviewing is more then simply providing a literature chapter, it is a process.
  • Reviewing the literature is a process
  • It goes on throughout any research, albeit in different forms
  • It must begin at the start, but will need to change in the light of changing research questions
Mapping relevant literature
In the early stages of research the task of searching for and reading relevant literature is primarily a matter of identifying what could be relevant, what significance it may have and how it might be used

There will be core literature that is very likely to be relevant, but also many other areas of literature, fairly indeterminate in character, that could be relevant.

Types of searching
  • Looking in catalogues, on library shelves, or in academic bookshops for relevant literature
  • Searching via electronic databases
  • Looking through or searching relevant journals, especially for review articles, 
  • Following up references in sources already found (this is a CORE option)
  • There is a tension between exhaustiveness and pragmatism
  • There is no perfect way to do this, there is always a trade-off. 
Types of reading
  • Skim reading (good to get an account of what is out there, the birds eye view)
  • Reading in search of specific kinds of information
  • Close or in-depth reading designed to understand and to assess the arguments and evidence put forward, and how these relate to the field of investigation (see e.g. hammersley, 1997)
  • Different reading strategies will be needed, at particular times, in dealing with particular articles and books, depending on the purpose they are serving. 
Writing a literature review
  • The purpose of a literature review chapter to provide a context and rational for the study
  • Avoid sequential paragraphs about each study: some degree of synthesis is needed
  • It is also important to evaluate the studies in methodological terms. Take care not to do this in a tendentious fashion: in other words, criticising those that don’t serve one’s purposes, while not critically assessing those whose finding some finds congenial or useful. 
Using the literature in other parts of a thesis, dissertation, or book
  • In an introduction (to help the reader understand what you are doing and why)
  • In a methodology chapter
  • In analysis chapter
  • In the conclusion
References
Hammersley, M. (1997) reading ethnographic research, second edition, London, longman
Hammersley, M. (2013). The myth of research-based policy and practice, London, sage.
Hart, C. (1998). Doing a literature review: releasing the social science research imagination, London, Sage.
McKinney, J. (1966) constructive typology and social theory, New York, Appleton-Century-Crofts
Oakley, A (2007). ‘Evidence-informed policy and practice: challenges for social science’ in Hammersley, M. (ed.). Educational research and evidence-based practice, London, Sage.

Questions:

How to combine interdisciplinary literature into a literature review
Thesis: is written for your examiners, and the examiners would be seen as the experts in their field, so their concerns and background will provide guidelines on what to write. On the other hand, some of the research you are conducting will be known to the examiners, but you need to write as though your readers do not know what they know. Supervisors help with that.

Question on emerging fields and lack of literature
What is published can be seen as literature, but also as data: if you have got people who play a role in the research context, then these people’s writing could be used to describe a wide spread view on the topic. Parallel fields and methods can be used to build the boundary for your research.