Marlot Kuiper

105 On Ethnography ing after its ‘implementation’. Rather, I ought to understand business as usual, and herewith encounter a ‘flux’ of apparently stable routines (Feldman, 2000). Studying processes however, is not easy. Change is something that occurs over time. Doing ethnographic research therefore takes time and effort. As I wanted to create the opportunity to see a variety of routine performances – that is both ostensive and performative aspects - and with this distil mechanisms that could lead to change or stability, I decided to go along with actors, interactions, and artefacts on the move, rather than staying in one place. This implies that I did not conduct a longitudinal study tracing how processes evolve over time in one particular operating theatre, but that I observed many checklist performances by various teams, in various departments. As said, the most important part of an ethnographic study is being present in the field. This ethnographic study is characterised by episodic observations, which means that I continuously switched between doing fieldwork and doing the analysis. The research sites were thus visited in various intervals and not continual. Conducting episodic observations had various advantages. First of all, the observations were typically data intensive. I produced a large amount of data in a relatively short time period. Moving from the field back to a desk during the phase of data collection allowed me to draw up and elaborate all the notes that I collected and start the analysis. Further, on the one hand negotiating acceptance and building relationships in the field is an important part of ethnographic fieldwork. On the other hand however, as a researcher I did my best to maintain the outsiders perspective and not ‘go native’. The episodic character of the observations made it more easy to balance socialization and objectivity. Data saturation determined the duration of the fieldwork. Though every performance of a routine is a unique instance in itself with unique features, I stopped the data collection at the point where I felt that data became repetitive and no new themes emerged. In conducting the observations, I used a shadowing technique (see also McDonald et al., 2005; Noordegraaf, 2000; Rhodes et al., 2007). By shadowing various professionals from both anaesthesia and various surgical specialties, I got to see the many routines that constitute professional work in the surgery department. I explicitly decided to shadow the various actors for full working days, since I was not interested in solely observing the performance of the SSC routine in isolation of other practices, but in painting a complete picture of the various professional 4