Dialogue in practice: the quality of conversation in professional groups with online support

This paper is concerned with the use of dialogue in the workplace, specifically in the formation of group-based, information-sharing exercises that involve the use of a technological Group Process Support System (GPSS). In a series of studies involving groups of General Practitioners in work-based settings, a system named TEAMWORKER (Gear & Read, 1999) was introduced to assist the course of a regularly-held meeting. The purpose of these sessions is to share information on individual progress, and to explore new ideas. The GPSS was employed to establish the range of opinion on various issues at practice meetings, with opportunities for free discussion and expressing changes of view if necessary. With information gathered throughout from sessions in this series, the quality of dialogue generated between participants is explored and analysed, to establish the links between dialogue behaviour and changes of opinion during the process. This may lead to a better understanding of how patterns of influence are operating in a group context with online support. In recent studies into the effectiveness of GPSS, the emphasis of research in this area appears to have favoured lab-based conditions (See reviews - Eom, 1999; Fjermestad & Hiltz, 1999). These studies have focussed on outcome- and/or participant-centred validation procedures (Gear & Read, 1999). A number of researchers have expressed the relevance of examining and modelling the 'process' of technologically-supported meetings in order to improve on the validity of GPSS in the workplace (Minkes, 1994; Eden, 1995; Finlay, 1998) in order to deepen our understanding of the effects of GPSS. In the present work, a field-based approach is employed to maximise realism and the accuracy of the data. Measures of process are the primary sources of information gathered during this research (i.e. an audio-recording of the group interaction, and a record of the individual inputs and group outputs via the GPSS during the session). This has enabled the development of a conceptual model of the influences (e.g. the maintenance of status and identity, group-based learning, etc.) and their effects on individual professional judgement during sessions of this kind. The study also compares findings with theories of group conversation, including a theory developed by Kantor (See Isaacs, 1999, Kantor & Lehr, 1975), which identifies four key aspects of the interactive process that should be present in order for group-talk to be transformed into effective dialogue, namely: moving (initiating action); following (agreeing with action); opposing (not agreeing); and bystanding (providing perspective). This 'Four Player Theory' also attributes corresponding qualities, or 'modes of behaviour' to each action that should be assumed to maximise the quality of dialogue. In the paper, theories are examined and applied to one of the GP group sessions held at a South Wales medical practice as part of this research. A number of questions emerge as a result of this exercise. While the GPSS assists in providing for the anonymous presentation of opinions, encouraging a more honest response by individuals, does this lead to more focussed interaction during periods of discussion? What are the main influences that affect the quality of dialogue during group meetings of this kind? In exploring these matters, this paper aims to build on current process understanding and accompanying theory.