In the traditional perspective of Grounded Theory (Glaser and Strauss 1967) as well as in its more organized and operationalized version (Strauss and Corbin 1990) the ruling idea is that theory emerges out of data analysis. Using extensive and intensive practices of analyzing the data from his or her field research the ethnographer ‘finds’ new theory. Academic discourse and existing theory play their part mainly in conceptualizing the ethnographic study and maybe in the form of “theoretical sensitivity” (Strauss 1994, 50) but the outcome of ethnography is said to be strictly routed in data analysis. Reflecting our own practices of producing ‘ethnographic knowledge’ we find that the well-known tradition of Grounded Theory methodology does not describe the process thoroughly. It does not take into account what we call the “double reference” of ethnographic analysis (Breidenstein et al. 2013/15): The development of analysis and of ‘theory’ in ethnography has to refer to the data and to the academic discourse at the same time. Practices of data analysis go along with practices of reading and discussing academic discourse. We are constantly testing if and how (potential) findings, categories, concepts or topics may fit to the academic discourse in the field: Which contribution could our research make? Can it tell (the discipline) something new? The occupation with theory does not only guide the process of ethnography in the beginning, it is a substantial part of the analysis all over the process. Theories have to be tested and qualified and used in manifold ways. The paper will put forward the idea of double reference of data analysis by giving examples from an ethnographic research process on assessment practices (Breidenstein/Zaborowski 2011; Breidenstein/Thompson 2014). Another problem one can have with the established framework of Grounded Theory is the notion of “data” itself: Of course we produce, handle and analyse “data” in ethnography, but these data are not ‘given’ but made (Emerson et. al. 1995). Producing and analyzing “data” are most important practices in ethnographic research and we should clarify what we are doing by analyzing our data. But data and data analysis cannot warrant ethnographic knowledge as such. The proof of ethnographic knowledge does not lie in the field or in the data at last but in the academic discourse which decides about the novelty and plausibility of ethnographic knowledge.

Breidenstein, Georg/Katrin U. Zaborowski: Disciplinary technologies and pupil redisposition: school equipment and homework diaries, in: Jeffrey, B. (ed.): Ethnography and Education Vol. 6 Issue 2 June 2011. Routledge (Taylor & Francis Group) (2011), S. 147-160 Breidenstein, Georg/Thompson, Christiane: Confession and subjectification in school performance evaluations, in: Fejes, Andreas/Nicoll, Katherine (Ed.): Foucault and a Politics of Confession in Education. Routledge: London/New York 2014, S. 25-34. Breidenstein, Georg/Hirschauer, Stefan/Kalthoff, Herbert/Nieswand, Boris: Ethnografie - Die Praxis der Feldforschung. Konstanz (UTB) 2013, (3rd ed. 2020) Emerson, Robert/Fretz, Rachel/Shaw, Linda: Writing ethnographic fieldnotes, Chicago Glaser, Barney G. / Strauss, Anselm L.: The Discovery of Grounded Theory. Strategies for Qualitative Research, Aldine Publishing Company: Chicago 1967 Strauss, Anselm/Corbin, Juliet: Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory, London: Sage 1990 (4th ed. 2015) Strauss, Anselm: Grundlagen qualitativer Sozialforschung, München: Fink 1994