Marshal B Anderson, M.Ed. Workshop 2 Co-operative Project, Spring 1999

The data used in this study is confidential at several levels and for several purposes. At the highest level it was an absolute pre-condition from the organisations involved that, if the data was to be used un-censored and beyond their control, the institutions represented must on no account be identified. At the whole-group level confidentiality within the institution was also a pre-condition. The reasons for both of these sets of pre-conditions may be seen as professional, defensive, or both, but they were acceptable for this study and cleared with the M.ED course tutor beforehand.

Intra-group confidentiality was vital from the point of view of the study. It is necessary that each group member be able to reply to the questions put with the confidence that they were able to communicate in a way that would make it impossible to be identified by the study author or any member of their establishment's administration. They also needed to be certain that no other member of the group would have access to their answers, a condition also necessary for the study.

This intra-group confidentiality was achieved by broadly similar means for the two groups. In both cases the groups assigned each member a number which they all knew so that they could refer to each other in the questionnaire but without giving actual names to the study author. The on-line group then set up anonymous e-mail accounts using several of the free services available so they could e-mail replies. The real world group were given stamped addressed envelopes to mail their questionnaires in. Once all the replies were received the numbers were randomly re-mapped so that no individual group member would be able to identify themselves, or any other member in the group in the final study. This method was outlined to the groups before they agreed to be involved in the study. It was also agreed that, once the data had been re-mapped, the original data would be destroyed.

Each student in both groups was given a questionnaire which asked for basic details of age, education, previous employment (in general terms, specifics were not requested) and their feelings about the current state of the group (see below). They were then asked to score each other member of the group on the following scale:

You will soon, or have recently, chosen to work in sub-groups from your class. Using the number identification system you have devised, score each other member of your class on the following scale:

1 and 2 for the first and second choices of the people you would most like to work with.

3 for people you feel you would have no problem working with.

4 and 5 for the two people you would least like to work with, 5 being your strongest rejection.

You may only use numbers 1,2,4 and 5 once. You may use 3 as for as many people as you like. If you don't feel you know a particular person well enough to make a choice, leave their box blank.

Each student was also given the opportunity to make any additional comments about the group making process they wished. The choices data is shown in appendix A.

At the point of this study the on-line group had been together for nearly a year, the real-world group for two terms, but the study refers to their social position after just one term. To try and get some measure of their position in group development a brief review of some major group development theories was carried out:

(Based on Jaques, 1984, page 35)

This table is based on what seems to have emerged as a consensus in group development theory; it provides a simple structure for the purposes of this study, but is by no means exhaustive. There is another, more recent body of theory that suggests these stages are either simplistic or simply wrong.

Accepting these ideas for this study. they seem to lead us to a generalisation that there are four stages of development plus a fifth stage where the group finishes its work. In referring to the whole of each group, each member was asked to answer the following question:

Some people think that a group goes through four stages of development; the first might be called formation and the last maturity. Which of the following statements best describes the current state of the whole of your learning group:

1)The group is still forming and we’re getting to know each other.

2) We know a little about each other; people are beginning to state their positions and opinions on more complex and/or contentious subjects.

3)The group is settled, people differ on some subjects, but we are generally tolerant at least of each other.

4)The group has reached maturity, we are working together and it works for all, or most of us.

The results this question gave were as follows:

Ideally we would have liked both groups to have felt themselves to have been at the furthest level of development as this would have removed a variable between them. However, the result is very firmly skewed towards the fourth statement, enough, for the purposes of this study at least, to give some degree of confidence that the groups have reached a fairly stable level of development.

As an aside, it is of interest that the real world group, who have been together for a shorter time, actually seem to have reached a greater level of cohesiveness or stability. It may be part of the nature of the different types of group that it takes longer for an on-line group to reach this level. This is discussed further in the conclusions.

It should be noted that the data collected from the real world group was retrospective - i.e. it referred to choices made in the past. While hindsight might have coloured these results, they should be fairly sound. The on-line group, however, replied to their questionnaires over a period of about three weeks while some were still going through the small-group forming process, this will tend to blur the picture. Morland, Levine and Cini (1993) suggest:

"…both the group and the individual engage in an ongoing evaluation of their relationship together, comparing its value to that of other relationships available to them."

In that respect it's possible that one group member might state opinions and preferences three weeks apart from another, by which time the relationship may have changed. At this stage in the study this flaw must be accepted, however, it is noted for future studies and was pointed out to the administrator of the on-line group as a refinement that should be applied in any further research.