Every methodology poses particular demands and grounded theory is not an exception. I strongly concur with the advice provided by Glaser (1978; 1998; 2001), based on his own experience and discussions with other grounded theorists, that the grounded theorist must:
tolerate confusion – there is no need to know a priori and no need to force the data;
tolerate regression – researchers might get briefly ‘lost’ before finding their way;
trust emerging data without worrying about justification – the data will provide the justification if the researcher adheres to the rigour of the method;
have someone to talk to – grounded theory demands moments of isolation to get deep in data analysis as well as moments of consultation and discussion;
be open to emerging evidence that may change the way the researcher thought about the subject matter, and be willing to act on the new evidence;
be able to conceptualise to derive theory from the data. This is perhaps the most important risk, as some people may experience difficulty conceptualising what is going on in the field; and
be creative in devising new ways of obtaining and handling data, combining the approaches of others, or using a tested approach in a different way.
Additionally, in adopting grounded theory methodology, the IS researcher has to confront two further risks. First, due to the minority status of grounded theory in IS research, it is likely that IS researchers, especially PhD candidates, will experience what Stern (1994), described as Minus-mentoring – that is, learning from books, employing grounded theory for the first time without the guidance of a supervisor with practical knowledge of the methodology. Minus-mentoring could result in methodologically unsound studies (Glaser, 1998; Stern, 1994). This has happened, for example, when studies claiming to be based on grounded theory have breached key tenets of the method (one of the main risks of using grounded theory within a second, overarching, methodology). However, ‘Minus- mentees’ can reduce this risk by (a) networking with IS researchers conversant with the methodology; (b) reading the ‘Grounded Theory Bibliography’ (Urquhart, 2001); and (c) participating in relevant discussion groups (e.g. IFIP WG8.2, the Grounded Theory Institute).[10]
Second, grounded theory seems to be easier to use when the researcher is sensitive, through having professional experience or knowledge, to the field under study (Glaser, 1978; Glaser, 1998; Glaser and Strauss, 1967). This sensitivity facilitates understanding or ‘verstehen’ (Weber, 1968). My substantial experience as a senior practitioner in the field of IS project management was a distinct advantage in eliciting information from participants in the same field. This experience facilitated the understanding of some of the more subtle issues in the study.
There is also the risk of finding something that is not new. What if this has been done before? This appears to be more a natural fear than a probable risk. To be sure, it is possible to study some emerging organisational phenomena just to come up with a theory that already exists in the literature. Yet this is unlikely. If the study is conducted as the method indicates, diligent researchers should have included the relevant literature (convergent and divergent) and detected variations and particularities. As Thomas Kuhn (1962, p. 30) said: ‘It is a truism that anything is similar to, and also different from, everything else’. A good grounded theory study should be able to point out similarities and differences, and to produce patterns that are particular to the substantive field of the research. Yet, as with any methodology, and indeed any human activity, there are no certainties.
Lastly, a grounded theory emerges through intensive intellectual action. Researchers need to interact with their data and while this interaction is often highly rewarding and satisfying, it is also extremely intensive, time- consuming and all absorbing, and the researcher must be persistent and resilient (as also attested by Urquhart, 2001).