Qualitative data analysis in CatchBob!
This afternoon, I tried to formalize a bit my current research approach to analyse qualitative data of CatchBob! The point is to benefit from users' annotations (in game) and the interview I conducted after the game (based on a replay of the activity). This leads me to the extraction of different valuable information that concerns coordination processes in the game.
This is based on Herbert Clark's framework of coordination (as explained in the book "Using Language"). In this context, coordination is a matter of solving practical "coordination problems" through the exchange of what he calls ‘coordination keys/devices’; that is to say, mutually recognized information that would enables the teammates to choose the right actions to perform so that the common goal might be reached. As a matter of fact, such information allows a group to mutually expect the individual actions that are going to be carried out by the partners. According to Clark, a coordination device is not only defined by its content but also by the way the persons who collaborate mutually recognize it. For that matter, Clark differentiates four kinds of coordination devices: conventional procedures (when a convention is set by the participants), explicit agreement (when the participants explicitly acknowledge the information), precedent (when a precedent experience allows participants to form some expectations about others’ behavior), manifest (when the environment or the information sent makes the next move apparent within the many moves that could conceivably be chosen).
This framework then leads to the creation of two coding schemes to analyze my data:
- What a participant inferred about his/her partner during the game. This coding scheme is clearly data-driven in the sense that it emerged from the players’ verbalizations (namely those extracted during the self-confrontation phase after the game)
- How a participant inferred these information about their partners: this one is theory-driven since I used Herbert Clark’s theory of coordination keys/devices to have clear categories about what happened
Now, there is another dimension that should be taken into account: TIME: different coordination keys are used at different moments in CatchBob, so I'm trying to put this together in a global model of spatial coordination. In the end, in the would express which kind of coordination keys are used to solve certain coordination problems in the context of a task mobile collaboration such as CatchBob. The potential outcome for this would be to understand whether specific tools can supports the coordination process (for instance would a location awareness tool be useful at a certain point the process').