Knowledge Management Capabilities Consensus: Evidence from a Delphi Study

Article excerpt

INTRODUCTION

The Delphi method, more and more known by mid- to long-term strategy development specialists, aims to collect via an open survey the justified opinions from a panel of experts in different spheres of activities. The procedure based on retroaction avoids confrontations between experts and preserves their anonymity. The results of a first questionnaire are communicated to each expert (including a summary of the general tendencies and particularities, opinions and justifications, etc.) that is then invited to react and answer to a second questionnaire developed in function of the first opinions collected, and so on until the strongest possible convergence of answers be obtained. The Delphi method distinguishes itself from usual group communication techniques on the following axes: (1) it helps to get experts opinions in a sector; (2) it allows to collect information at a distance, via the Web, fax, or mail, while the respondents have not to meet; (3) it facilitates the task of identifying and selecting experts since the number of experts participating in the study is limited to 7 to 18 (Paliwoda, 1983; quoted in Okoli & Pawlowski, 2004, p. 4); (4) it is flexible in its conception and its survey administration: this allows rich data collection leading to an appropriate understanding and a consensus on knowledge management (KM) capabilities; (5) it allows quasi-certainly to get a consensus via the issuance of consecutive questionnaires; (6) it allows a controlled feedback consisting of a series of steps from which a summary of the previous step is provided to the participants; so, if they want, they can review their previous judgements; (7) it has an advantage over other methods of group decision making given the analysis of anonymous experts opinions which are identified before the study (for example the nominal group and the analysis by social judgement) (Rohrbaugh, 1979; quoted in Okoli & Pawlowski, 2004, p. 4); and (8) it can be used as successfully in management, economic, or technological sectors as it can in social sciences.

On the other hand, several constraints are limiting the use of the Delphi method: it is lengthy, costly, fastidious and intuitive rather than rational, among others. In addition, the procedure constraints (multiple rounds of surveying) are questionable since only the experts that stray from this norm have to justify their position. However, we can also consider that strayed opinions, in prospective terms, are more interesting than those close to the norm. Finally, the interactions between the different hypotheses proposed are not taken into account and they are even avoided, leading the promoters of the Delphi method to develop probabilistic cross impact methods. Furthermore, it is obvious that a Delphi study does not base itself on a statistically representative sample of a population. It is rather a mechanism for group decision making that requires the participation of qualified experts having a clear understanding of the phenomenon being studied. It is for this reason that one of the most critical factors of this approach is to find qualified experts.

The objective of the present research is not to conduct a study aiming the improvement of the Delphi method or to perform a comparative study between the Delphi method and the other methods or techniques of group communication, but rather to try to understand the phenomenon of KM capabilities by identifying the key concepts and the characteristics of each concept in order to reach a consensus from experts. Thus, the use of the Delphi investigation shed the light on the ideas of KM experts and helped to find a consensus from experts on the theoretical concepts of the KM capabilities.

The paper is structured as follows: first, a brief history and some applications of the Delphi method in management, information systems (IS)/information technologies (IT), and KM are presented; second, the steps of the Delphi investigation related to this study are described; and the paper ends by providing some examples of general comments from the experts participating in the study and their profile. …