Implications of Using Software to Support Qualitative Research

Abstract

There is a long-standing debate about the appropriateness of different research paradigms and methods for studying organizations. This is taken up in the Strategic Management Journal (Rouse & Daellenbach, 1999, 2002), in a series of special issues of MIS Quarterly (Orlikowski & Barley, 2001) and in papers in other journals (Mackenzie & House, 1978; Meredith, Raturi, Amoako-Gyampah, & Kaplan, 1989; Modell, 2010; Pfeffer, 1993; Richards, 2002; Richardson & Robinson, 2007; Scapens, 2008; Vaivio & Siren, 2010). Some argue for a more in-depth approach to research using qualitative and interpretive methods. The outcome of the methodology debate is an increasing acceptance of qualitative methods. In the past, qualitative research, employed manual techniques for processing data. Researchers were the principal actor in coding, sorting and interpreting data, though some research assistance, often unacknowledged, may have been employed. However, along with increased acceptance of the use of qualitative research has come development of tools that automate various aspects of qualitative research activity. Yet, the use of computer-assisted qualitative data analysis software (CAQDAS) by qualitative researchers remains contentious (King, 2010) and little discussed. This article argues that use of qualitative research tools can affect the outcomes of research projects in ways that researchers may not have considered, and the findings from research where such tools have been used may be qualitatively different from findings derived from research where no tools have been used to automate coding and analysis processes. Such differences may be due to a capacity to conduct complex analyses more easily (Hutchison, Johnston, & Breckon, 2010); to manage very large data sets (Holstein & Gubrium, 2003); or they may be due to unanticipated processes (Davidson & Skinner, 2010), unnecessary complexity, or data corruption. There is a need for critical discussion about the impact of CAQDAS, and its potential to affect - possibly detrimentally - research outcomes (Blismas & Dainty, 2003).