Exploring actor-network theory and CAQDAS : provisional principles and practices for coding, connecting and describing data using ATLAS.ti

Wright, Steven Timothy (2016) Exploring actor-network theory and CAQDAS : provisional principles and practices for coding, connecting and describing data using ATLAS.ti. In: ATLAS.ti User Conference 2015 – qualitative data analysis and beyond :. Universitätsverlag der TU Berlin, Berlin. ISBN 9783798328228

Abstract

This paper explores how ideas from actor-network theory (ANT) can be drawn on to inform ways of using computer assisted qualitative data analysis software (CAQDAS) in an ANT-oriented project. Through this it explores some of the challenges ANT poses to conventional uses of such tools, and the resources ANT provides for re-considering their agency in research practices and possibilities for future developments. CAQDAS is often associated with particular approaches to engaging with data (e.g., coding data and retrieving the codes, abstracting and reducing data to themes etc.). These approaches have become dominant enough such that they are often presented and/or interpreted as the right, or only way to work qualitatively with qualitative data. In opposition to this orthodoxy some orienting principles are proposed from the ANT literature along with its intellectual antecedent ethnomethodology. The proposed principles are: freedom of movement and data, logging the inquiry using Latour's four notebooks, coding and following heterogeneous actants as cases, supporting contextual exploration of fluid and multiple ontologies, staying close to the words of the actants and working in a scale-free manner that enables shifting magnifications and assemblages to preserve detail rather than abstract it into themes. A final principle concerns the intentions of ANT-informed approaches to assemble a detailed description, which are contrasted with the intentions of approaches aligned with Grounded Theory to abstract data in order to construct an explanation. These principles are explored and illustrated with detailed descriptions that draw on examples from a multi-modal ethnographic PhD research project. The project used heterogeneous data to explore the information infrastructures and classification systems used in craft beer judging. Examples of how that diverse dataset was coded and connected are used along with excerpts from a reflective journal of the struggles and ideas for using CAQDAS to illustrate ways of effectively using ATLAS.ti in ANT-oriented research projects.