I am going to have a stab at summarising what I suspect are the main 'messages' that have come out of a number of detailed postings that make up the threads relating to meta-narrative reviews (MNR). I hope my efforts to condense (and relative ignorance about MNR methodology) will not do an injustice to the depth of some of the discussions. If I have, do tell me and correct any of my errors or omissions.
UNIT OF ANALYSIS
Questions were variously asked about what the differences were between 'story line' vs. 'narrative' vs. 'meta-narrative' vs. 'discipline', vs 'tradition'. And seemingly the implicit question was which one should I use to analyse the data with?
MNR aims to make sense of diverse literature where concepts have been defined and researched differently because of the researchers' accepted practice and beliefs. This is where Kuhn comes in with his concept of a paradigm, but a finer grain 'unit of analysis' is deemed to be more useful as within a paradigm there may be more than just one set of practice and beliefs. Experience to date with MNR does seem to indicate that the 'tradition' is a useful unit of analysis.
A research tradition is defined as defined “as a coherent body of theoretical knowledge and a linked set of primary studies in which successive studies are influenced by the findings of previous studies”
(I realise Ihave not 'defined' the other terms. You probably know this, but definitions may be found in Greenhalgh et al.'s 'Storylines of research in diffusion of innovation: a meta-narrative approach to systematic review' Social Science & Medicine 61 (2005) 417–430 and 'Tensions and Paradoxes in Electronic Patient Record Research: A Systematic Literature Review Using the Meta-narrative Method The Milbank Quarterly, Vol. 87, No. 4, 2009 (pp. 729–788)' has a Glossary).
A question that arose was how many meta-narratives need to be mapped. The answer given was more than one, because if a concept was defined and researched in pretty much the same way by everyone, then a MNR would probably not be the ideal method to use.
As traditions evolve and change, one assumption was that there was some means of communication within and between traditions. It was raised that this may not necessarily be happening and questions arose about whether it was possible to map such talk (or its absence) and/or what the implication was to the mapping process if none of this 'cross talk' occured.
THE SOCIAL DIMENSION and FOUCAULT
Whilst Kuhn suggests that knowlegde is a social construct, one weakness raised in his writings is that of the influence of the social dimension on knowledge construction. Foucault's knowledge-power theoretical lens could provide an alternative way of viewing data... does anyone feel like operationalising this into a review method? Is there a need?
Two other points came up - middle-range theory and software. as these are more 'generic' issues to both review methods I have started separate threads on each.