9312
Group based research: commitment or compromise?
Geoff Anderson, University of Technology, Sydney
Abstract
Focus group interviews can provide the researcher with detailed and carefully considered qualitative data from the people most knowledgeable about the research issues. Group interviews are particularly useful when time and resources are limited or when participation, consensus and a degree of commitment need to be built into the research process. This paper considers some of the special characteristics of a focus group and discusses the role of a facilitator who is also a researcher. In particular it examines ways to ensure group consensus does not exclude individuals who may feel compromised rather than committed if the final outcomes fail to reflect all perspectives.
Introduction
In a bizarre little experiment conducted in 1969, psychologists measured the performance of cockroaches running away from a light source. They found that the cockroaches ran faster if the runway was lined with an ‘audience’ of fellow cockroaches, each watching from a perspex box [1]. In other words, the presence of a group had a facilitating effect upon performance. As a researcher who works largely with groups I found this result strangely comforting. It seemed to provide some sort of natural vindication for my methodology.
This is not to liken group participants to cockroaches. The vast majority of participants have nothing at all in common with this creature apart from their willingness (usually) to make an extra effort in the presence of their peers. It is said that participants in a group are both audience and co-actors: they perform and evaluate at the same time. As audience they retain an individualistic and critical perspective; as actors they must take their cues from others. A question for group facilitators is whether this leads to an inherent role conflict. Is the individual’s commitment to the group’s decision lessened in any way through a need to compromise his or her own position for the sake of consensus?
Although consensus is widely regarded as the best way of achieving a quality group decision, a false consensus (one which is more apparent than real) may later prove counter-productive and could easily throw into question the legitimacy of the final outcome. The aim of this paper is to make some observations on this matter of consensus and to consider why groups, despite their limitations, can still be a viable option for data collection.
Why use groups?
As any teacher or committee member knows, groups can be exasperating, difficult and inefficient. To use a group as a research tool seems to be asking for unnecessary complications. Yet I firmly believe group based research methodologies offer a means of information gathering well suited to a field such as Adult Education, where an acknowledgment of the rights of people to participate in decisions which affect their lives is a dominant value. In fact the need to build consultation and consensus into the data collection process first led to my involvement with group based research, in particular the single-purpose focus group.
While focus groups are normally associated more with market research than academic inquiry, in recent years there has been a growing interest in qualitative methodologies which include a group interview component[2]; A focus group consists of people with particular knowledge and experience brought together to represent the views of a wider constituency. For instance, ten supervisors might provide data or ideas on behalf of middle management in a particular industry. The researcher’s emphasis is on gathering data, opinions and ideas quickly and accurately. In my field (HRD) this information is usually gathered for purposes of occupational or competency analysis, training needs assessment or program design. I have also used focus groups to explore attitudes to change, feelings about work and to generate ideas about specific problems. The purpose, therefore, normally dictates a fairly structured, instrumental approach and most of the research questions will have been formulated before the group convenes.
In this sense the type of research I am concerned with is quite different to participatory action research, which aims to effect organisational change or political empowerment through group reflection and action. But as with action research, the focus group facilitator can avoid the stance of ‘outside expert’ which is often resented by people within an organisation. My role is to acknowledge and uncover the expertise within the group, guiding the discussion without actually contributing to its content. For most trainers and educators, leading a group discussion is familiar territory, hence another reason for my belief that such an approach is highly suitable to research in these areas.
Other advantages
There are several additional advantages which I see for group based methodologies. In contrast to an individual interview, a group interview allows interaction, debate and varied perspectives to be brought to bear on the topic. The result is data which is qualitatively different to that which could be obtained by summing the results of individual interviews.
There is also a large body of empirical evidence to suggest groups can recall information more accurately and more completely than individuals in isolation[3]. In my experience group membership also has a tendency to reinforce individual commitment to the group’s norms and decisions making subsequent implementation easier. Groups themselves promote commitment, provided, of course, there is no feeling of personal compromise.
The group processes
A focus group, brought together for a short time to address a specific issue, is task oriented and works to a set agenda. This helps obviate dysfunctional power games and time wasting diversions. I never attempt to facilitate a focus group without some tools to provide structure and closure. Three in particular are most useful in my area: Nominal Group Technique (NGT), Dacum and Functional Analysis. All provide a formal framework which makes efficient use of time and encourages a participatory climate
In each of these processes the researcher is also the facilitator. The objective is to elicit information and ideas previously unknown, unconnected or unavailable. Using the experience of the group, the facilitator poses a series of questions which all members attempt to answer. This may be achieved through silent reflection, followed by a sharing of ideas and a subsequent evaluation and ranking (NGT) or through a logical analysis of the component parts of an entity (such as an occupation) after an initial brainstorm to generate ideas (Dacum and Functional Analysis). In any case, each session has a definite objective and follows a planned agenda.
A collection of people or a group?
The fact that certain people are in the same room at the same time does not make them a group. A focus group arrives with no history and leaves with no future. As Shaw[4] points out, such a group has no traditions to draw upon and there are no expectations of future interaction. I prefer, where possible, for group members to be drawn from a variety of sources. However this means the members are usually unknown to each other and may feel little responsibility for any consequences which might follow from their deliberations. This presents the first challenge, since I believe a vital part of the facilitator’s role is to give the group a very strong feeling of responsibility and a real sense of purpose. Later commitment will depend upon this.
Recent thinking suggests cohesiveness is a major defining characteristic of a group[5]. Perhaps the easiest way of turning a collection of people into a cohesive group is through giving each individual a stake in a common goal. I have seen that a group of strangers can work effectively as a team and establish friendly interpersonal relations in quite a short time provided they have a sense of some common purpose and a structure is provided for the group processes. A focus group also benefits from a slight sense of urgency and an awareness of the relevance of their work. Personal concerns may need to be subjugated in order to achieve a specific task. Commitment to outcomes may then be built on the basis that all participants have contributed equally and the final product is the sole reason for the group’s existence in the first place.
Since they are short-lived, focus groups seldom suffer from the tensions which can arise when task is continually stressed over relationships. With NGT discussion is limited to certain times and themes, mainly for purposes of clarification. This is useful for preventing arguments developing. The fact that all contributions become group property once they have been listed and quantity of ideas is emphasised also lessens the need for individuals to defend a position.[6]
With the right choice of participants, facilitating a focus group can be an enjoyable and productive experience. Unfortunately, choice of participants is something over which facilitators may have little control. We ask for experienced, representative and articulate volunteers. We get whoever can be spared on the day. We ask for ten participants, we are sent four. There may be other problems too, particularly if individuals feel they are there to argue for a position or there are disruptive, taciturn or unmotivated participants within the group. Since researchers seldom have authority in these matters we rely very much on the goodwill of supervisors to send suitable people and on the willingness of the people sent to cooperate with us.
With less than ideal participants the facilitator must work harder and subsequent validation of results becomes more important. In any case, a questionnaire or interview follow-up is important to allow those not in the group to comment and so to become involved in the project. Data provided by a group is useful only insofar as it reflects generally agreed opinions within the group’s constituency or takes account of all the situational variables. The researcher should normally seek to validate such data. The fact that the results of the workshop will be validated actually assists the group achieve consensus, since their collective decision is largely provisional.
Seeking consensus
All interactive groups operate on the basis of consensus, an agreement to accept the norms and decisions of the group in exchange for membership rights. Fear of rejection can be a strong incentive to feign agreement. Still, a large group will normally take much longer to reach consensus than a smaller one. With Dacum, which requires all participants to agree before moving on, consensus is vital. Thus having co-operative participants willing to concede the occasional point is important. Although NGT is not in itself a consensus-seeking technique it is preferable for all participants to feel comfortable with the final decision, even if they personally do not agree with it.
Consensus is neither a compromise nor a democratic vote. Compromise involves finding a middle path between two positions, while consensus means a willingness by all parties to accept the group’s decision whatever that may be. A simple majority vote results in winners and losers, a situation hardly conducive to later commitment. Sheer fatigue may lead to apparent consensus and the facilitator needs to continually test consensus through questioning and sensitivity to body language. Given that unanimous agreement is not always possible, even when discussing ostensibly factual data, I feel the best way to achieve consensus without compromise is to agree to a provisional position. In this way those not in agreement can at least agree to reserve their final judgement until the data has been further validated. Their integrity remains intact while the need for a decision has been met. The researcher, concerned with gathering data or generating ideas, is looking for broad agreement rather than total assent in every case.
Some reservations
Highly structured group processes work best when time is limited, there is a specific task to be accomplished, participants are willing to share ideas and the facilitator is accepted as the group leader. These conditions may not always be present and in an unstable organisational environment or when dealing with highly individualistic participants a less formal, more open procedure is indicated. The restrictions imposed by the structured techniques may be uncomfortable for those who wish to discuss things more freely or who feel inhibited by predetermined group strategies.
Additional problems may be created by unbalanced representation, by the facilitator’s failure to separate facts from feelings or by the group’s unfamiliarity with certain issues and situations. Since the group is only short-lived a lack of commitment on the part of individual members may indeed lead to a false consensus and later feelings of personal compromise. The careful selection and briefing of the group prior to the workshop and the ability of the facilitator to recognise such problems can, however, go some of the way towards preventing this. It is also difficult for a group to provide substantive or quantifiable data. In fact several workshops with the same or different groups may be needed to develop a really comprehensive picture. The phenomenon of ‘group think’[7] is also a reality with a small group of similar participants. Groups are a powerful research tool but all the normal cautions in regard to data collection still apply.
Conclusion
People who give of their time and energies to contribute to a group interview reasonably expect to have their voices heard. As facilitators we do our best to ensure everyone has an equal say, group harmony is maintained and the coffee arrives hot. A focus group can be tiring and frustrating. However, I worry only if it ceases to be a group: if individuals start taking sides or hidden agenda emerge or political issues dominate the discussion. At times like this the only recourse is to call for time out or even reconvene a new group at a later date.
I have found participant satisfaction with the process is usually high. They may be surprised at the amount of data they were able to produce or the richness of the experiences they had to draw upon. While individuals may hold differing views a formal technique, such as Dacum or NGT, allows them to feel they have contributed to the final group decision. In any case, if strong differences exist traditional research methodologies would do no more than reveal them. In a group situation there is a chance, at least, of moving towards a resolution.
The techniques I have referred to in this paper were designed for special purpose applications but need not be limited to these. Many teachers use a version of NGT for class brainstorms, while the others simply provide a methodical way of developing, clustering and organising data, for whatever purpose. They can be used both to stimulate creative thinking and to tackle a problem in a logical manner. Hence their versatility for a range of situations and issues, particularly at an organisational level. But it is their participatory nature which I feel makes them worthy of serious consideration by those of us involved in research in an area where people’s ideas and opinions are seen to matter.
[1] Zajonc 1996 cited in RS Baron, N Kerr and N Miller (1992), Group Process, Group Decision, Group Action. Buckingham: Open