The methodology that we will be using for the Affordances for Learning project is based on Actor-network theory and Complexity theory. One of the key issues, both methodologically and epistemologically, is obviously: how is the data constructed, and how do we make sense of it?
There’s some interesting stuff in there …
In traditional research design, you are required to specify your methodology and your ‘data collection’ in some detail, in advance, at the proposal stage. However, firstly, its more useful if we are looking at complex adaptive sytems, or complex adaptive networks (CANs) to see the data as ‘constructed’ rather than just ’sitting there’ waiting to be ‘collected’.
Secondly, it is useful to limit the extent of prospective coherence in our methodology, and focus instead on retrospective coherence. If we define affordances as the product of the interaction between the individual and the environment, then affordances are what emerges after, and through, such interactions, and affordances are subject to the changes in the identies of the participants and the micro-ecologies, both of which are self-organising to at least some extent.
If meaning is contextual (broadly speaking it is, as a working hypothesis) and if the context, or the ecology, including the identities of the actors within it, is changing, then in any Complex Adaptive Network we really do need to focus on retrospective coherence. There is no reason why we should not specify what coherences we might find, prospectively, but we should not be limited to that, or by that.
Rather, we should set up our research so that we are confident that several ‘tracks’ of data (or sets of ‘traces’ in Latour’s terms) are generated during the period we are interested in, and then we should retropectively explore those traces (preferably with the help of the people who constucted them) to see if there are any emergent events, patterns, identities, activities, etc that we can identify and describe.
What this does is it separates off the generation and construction of primary data from the analysis. Ideally the data should be generated naturalistically, within the process as it would happen without the research. The process of sense making should be able to engage retrospectively, without privileging any specific aspects ahead of time - without giving any parts of the data a ‘heads-up’ so to speak.
Our curiousity, in other words, should not be confined to either confirming or rejecting the null hypothesis, we should look for the unexpected and the surprising, and systematically explore whether we can make sense of the unexpected, based on the traces – the ‘data-within-narratives’ – including both the ’stories’ and the ’story-tellers’ and the interactions between the two.
