What contextual knowledge do re-users of qualitative research need to access in order to make sense of the original data?

Let’s begin by drawing a distinction between two kinds of context:

1. Descriptions of the substantive context of a study - the myriad kinds of knowledge about the local field of study that the originator needs to draw upon in order to interpret and analyse the data.

2. Descriptions of the methodological context - the guiding frameworks, means and processes through which the study was conducted.

This section of the guide explores both these two kinds of information, taking issues of methodological context first and substantive context second.

In relation to originators’ work, it is clear that only substantive context is likely to be fully analysed and documented in a routine way, since it is something that they need to examine in order to make sense of their data. Methodological context, on the other hand, is less likely to be fully documented or examined.

The extent to which methodology is documented in an original study will depend on how reflexive originators are being about their methods. It is possible that a detailed ‘warts and all’ account of methodological decisions and reflections is made available for deposition, but it is more likely that methodological information will be confined to standard, ‘official’ descriptions of the project’s methodology of the kind provided in final reports. These often read along fairly well established lines (such as defining the field, sampling, negotiating access, gaining consent, and so forth). It is clear, however, that these ‘public’ accounts may provide little in the way of insight into the messy actualities of fieldwork and research relationships. Yet this kind of detailed methodological information is of considerable value to re-users; in fact, we would argue that it is essential.

For example, it’s clear that methodological context is not only about the whys and wherefores of technique and process, but also about the case-specific and complex relationships that are negotiated between the researcher and the participants during the course of research. These include consideration of differentials in power and status, the contingencies of planned and chance interactions, along with many other variables. When we archive our data for others, we cannot ‘account for’ this necessarily complex and particular aspect of context in any simple or straightforward way. We cannot, certainly, wish it away, for the quality and nature of the dataset is to a large degree dependent on its specifics. That is, any qualitative data-set will necessarily be generated within the parameters of particular research relationships, such that the data cannot be seen as detached from or independent of them.

All of this suggests that context does not comprise a set of static circumstances that the originator is ‘surrounded by’ in the course of study. Instead, as Holstein and Gubrium (2004) argue , context can be better thought of as ‘a fluid, socially emergent constellation of contingent factors that are “worked up” – not just encountered - in the course of everyday interaction’. All originators can do is to ensure this elasticity is ‘stopped in its tracks momentarily to allow for description and analysis’ (ibid. 309), through providing a rich account of relevant factors such that subsequent re-users can understand the conditions under which the data were generated and the conclusions drawn. As we will show in what follows, the existing guidelines on providing documentation do not in our view recognise this fluid quality of context sufficiently clearly.

There appear to us to be three major issues that questions of context potentially introduce into the debates on re-use:

1. How should we ensure that we provide an appropriate, reflexive and useful representation of the project’s methodology?
2. How should we document the wider, substantive context of the dataset – such as information of a socio-historical nature?
3. To what extent should we take re-use into account as we are generating our data and assembling the data-records?

The first two questions both regard documentation – how to ensure that the necessary kinds of knowledge and information are communicated to subsequent re-users. The third one concerns methodological process itself and opens up a number of quite contentious issues.

Next-->