Case Study 6: Educational Psychology

Introduction

The University of Strathclyde MSc in Educational Psychology is a two-year training course accredited by the British Psychological Society. Incoming trainees (11 females and 1 male for the cohort monitored here) are all psychology graduates with three to four years' experience of work with children, via research or teaching. The course involves lectures, seminars, tutorials, both group and individual project work, and practice placements. The trainees attend the University on two days per week for the lectures, seminars and tutorials, and spend another two days per week on individual placement in Psychological Service offices in central Scotland, carrying out professional work under the supervision of "practice tutors". Project work is organised by the trainees themselves.

Since for much of the week the trainees are in separate locations, in the past if they needed to collaborate (e.g. over seminar preparation and group projects) or obtain guidance from course tutors, they had to meet whilst in the University, despite having a full schedule during this time, or else make do with telephone calls. For this reason, it was decided in 1997 to introduce an email and conferencing system that would allow trainees to communicate with tutors and each other without having to worry about constraints of time and place.

Characteristics of the Communication Technology

Access to this system, which operated from networked machines in the Psychology Department at Strathclyde and modem links in Service offices (and/or at home in some cases), was provided to the trainees from early 1998. Internet resources, including Netscape and a course web site, were made available at the same time. The conferencing system used First Class 3.5, which allowed the trainees to post messages to either a shared discussion space or personal mailboxes, and also provided a synchronous 'chat' facility. This software was chosen because both it and connections to an appropriate server, run by the Scottish Council for Educational Technology (SCET), were already available in Service offices for eleven of the twelve trainees on the course. It also had a relatively transparent graphical user interface and a range of menu-driven facilities, important considerations given that only three of the group of twelve had used email previously at all.

Stated Purpose of the Conferencing

In October 1997, as a first step towards the introduction of the First Class system, examples of email and conferencing facilities were demonstrated to the new intake of trainees, and possible advantages of using these facilities for course communications were discussed. Those specifically highlighted were:

• benefits for course administration and group project work in terms of ease of communication
• the capacity to generate jointly written material, including that needed for presentation at an annual professional development conference
• the possibility of contact with trainees on a parallel course at Dundee University, and with other professionals
• the chance to gain experience of the use of resources of this kind

It was also mentioned that there was a need for qualified trainees who would actually make use of computer conferencing. Despite Scottish educational psychologists having been provided with a dedicated conferencing system, thus far the take-up by existing professionals had been poor, and the Scottish Office wanted to see an improvement on their investment.

After this demonstration, the introduction of a conferencing system into the course was mooted. It was pointed out that it would be desirable to monitor the usage of that system, in order to establish whether it was helpful, and how it might be improved for future trainees, but it was stressed that any intrusion and extra work involved in making monitoring possible would be kept to a minimum. In particular, online messages would not be inspected without consent. The trainees expressed concerns at this stage about use of the conferencing system involving extra work on top of an already heavy load; and conversely, questioned whether the group would actually find things to use the system for, and if they did, whether it would benefit everyone. Despite these reservations, though, they agreed to proceed.

Stated Principles of Use

After the trainees had agreed to proceed, a "closed" discussion group, accessible only by the trainees, course tutors and project consultants, was set up on the SCET server. Training in the use of First Class was then provided within the Department, backed up by a set of paper-based guidance notes. This training focused on sending messages to the shared conference space and private mailboxes, and on replying to these. No introduction was given to the use of the synchronous chat facility, because it seemed likely to present a confusing experience for new users. Trainees were asked to check their connections to the conference from their placement offices, and were offered help, where necessary, in installing software and setting up modem links. Due to networking problems, field connections in fact proved difficult to establish, and five weeks after the system had been set up, only eight of the twelve trainees were fully online. Two of the remainder continued to have problems for five weeks after that.

While these problems were being attended to, those who could access the First Class facilities were left to make informal use of them for contacting course tutors and each other. As things began to settle down, however, the trainees were given a more specific incentive to become familiar with using the system. One month after it was set up, they were divided into two groups of six and set the task of collaborating within their groups in whatever way they chose to produce a seminar paper on a set theme, making use of email where appropriate. For each theme, details of three key readings and four to five 'starter' web sites were made available by tutors. Groups were to review the materials relevant to their theme and produce a short briefing paper containing a summary and critique of previous work, and guidelines for professional practice. This paper was to be discussed at a face-to-face seminar. Groups were allowed approximately five weeks to prepare their paper, and worked consecutively, with the first group (Group 1) making their face-to-face presentation one week after the second (Group 2) had been set their task.

In addition to the seminar task, the trainees were encouraged to use the conferencing system to collaborate on their research projects, and on the preparation of a poster for a professional conference later that year. No restriction was placed on less formal communications. During this initial period of its operation (Spring 1998 - Autumn 1998) the trainees did not use the system (or any other) for external email contact: exchanges with tutors and between themselves were the only uses they made of email throughout this time.

From Autumn 1998 onwards, at least some of the trainees began to make use of their private mailboxes for external email contact, and also to try out other email systems (primarily the generic Pegasus system provided by the Psychology Department, operating via Departmental and University servers). Some also began to explore use of the First Class synchronous chat facility. In addition, the shared conference space was divided into two distinct areas, one (a continuation of the existing conference) reserved for general issues, and the other for discussion of professional/work-related topics. Access to both of these conferences, and to individuals' private mailboxes, was maintained after the trainees completed the course in Summer 1999, and moved out full-time into Psychological Service offices as probationers, although monitoring of usage of the system for the purposes of the present research was ended at this point.

Character of the Communication

In general, the trainees made little explicit comment on the nature of the communications afforded by the conferencing system, perhaps because so many of the group had not used email previously, and so had little to compare their experiences to, other than face-to-face interaction. The system was in fact seen as imposing some constraints on exchanges, but less because of its inherent characteristics, and more because of difficulties accessing it on a regular basis, due to either limited resources (see below) or problems finding the time to do so.

There was in addition some limited awareness that email might not be the best channel for more complex or sensitive interactions because of the extent to which it depersonalised exchanges and undermined genuine interaction.

What was more apparent was the emergence of clear patterns of usage of the system in terms of both the function and frequency of messages. These were shaped in part by the constraints noted above, but information from the various available sources (message archives, logs of activity completed by trainees, interviews with trainees and course tutors) all pointed to the seminar task as having had the largest influence on how the conferencing system was used, both at the time and subsequently.

Three methods were used to examine the message archive and determine patterns of use of the conferencing system during its first five months of operation (mid-February to mid-July 1998):

• a simple count of the frequency and timing of online messages, and their relation to the tasks which had been set
• an examination of the relative contribution of different course members
• coding of the function and length of messages

Looking at message frequency, a count of messages per week after start-up showed a small flurry of activity for the first four weeks (on average about five messages per week were sent to the shared conference space), as those who were able to gain access tried out their connections. This subsided to an average of one message per week for the next four weeks, despite Group 1's seminar task having been set, although this was explicable in part by the period coinciding with the Easter vacation. Over the next four weeks message frequency climbed sharply to a peak of 20 per week, as the remainder of the trainees established connection, Group 2's seminar task was set, and Group 1 finalised their paper for the face-to-face seminar. The subsequent four week period, leading up to Group 2's face-to-face seminar saw a drop in frequency to around three messages per week. The period after that, leading into the summer vacation, saw a further decline to, on average, one message per week. At no point, then, was message frequency particularly high, but this was not surprising, given that the trainees had much else to do (see above). Of greater significance was the fact that there was a clear coincidence between on-line activity and the group seminar tasks, especially in the periods immediately prior to the face-to-face sessions, although this trend was more marked for Group 1 than for Group 2. Thus, whilst it was not necessary to use email for the seminar tasks, the indications were that it had proved helpful to do so, and for this activity more than any other.

Moreover, whilst messages were sent as early as 8.47 a.m. and as late as 10.00 p.m., the vast majority were sent during normal working hours, exactly the time that had been difficult to use for exchanges of this kind before the system was in place. There was some skewing of traffic towards certain days of the week, with most messages sent on Wednesday, Thursday and Friday, and relatively few on Mondays and Tuesdays. The latter were however the two scheduled days in the University, when, as noted previously, there was little free time available, and what there was permitted face-to-face contact.

The data on the relative contribution of different members of the course both confirmed and qualified the impact of the seminar tasks. In the first place, use of the shared conference varied widely across the trainees, from no contributions at all in two instances, to a maximum of 24, with an average of 6.5. The two course tutors made infrequent contributions (6 each on average over the monitoring period), but since the system was intended to be primarily for the trainees' usage, they restricted their contributions to important points of information and responses to direct queries.

Overall, even taking initial connection problems into account, only five of the twelve trainees could be classed as regular users (i.e. those who made a reasonable number of contributions, spread over a period of time). However, the number of contributors did increase in response to work on the seminar tasks, although there were differences in this respect between Group 1 and Group 2. For the former, four out of the six group members became regular contributors for at least the period preceding their face-to-face seminar; for the latter it was three at most, and then much more briefly.

Coding of the function and length of the messages to the shared conference space helped to fill out further the picture of the on-line interactions. All messages during the period inspected were identified as falling into one of nine possible categories:

• connection and hardware or software issues
• contributions related to seminars
• tutor contact and advice
• peer contact and advice
• social exchange
• project work
• conference preparation
• course administration
• professional issues

This data confirmed the impact of the seminar tasks on system usage. During the eight week period covering the lead up to the face-to-face seminars for Group 1 and then Group 2, there were only two weeks when communications relevant to the seminars were not amongst the most frequent categories of message. Average message length also jumped during this period from 1K to 8K.

Two further points are relevant here. The first is that closer examination of the seminar-relevant messages showed them to be of two kinds: brief exchanges of information, or lengthy attachments of drafts of sections for the seminar paper with requests for feedback (hence the jump in message length). The following are typical instances of a brief exchange and a request for comments on an attached draft, respectively.

Thus whilst use of the conferencing system did seem to have been particularly focused by the seminar task, this usage in no respect paralleled face-to-face style discussion. It was, however, perfectly appropriate for what the trainees needed to achieve in terms of joint production of a seminar paper, especially since they did have the opportunity to meet face-to-face if discussion was really necessary (see 'Getting things done' below for further thoughts on this).

The second point is that whilst Group 2 made fewer on-line contributions during the weeks preceding their face-to-face seminar, the characteristics of the messages they did send were similar to those sent by Group 1 during the corresponding period. In other words, email was serving the same type of function for both groups, if not to the same extent.

The differences between Groups 1 and 2 in their degree of involvement in the shared conference during the seminar task contrasted markedly with the more intensive and evenly distributed participation observed in the trainees' face-to-face seminars (see 'Getting Things Done'). However, these differences were not in fact a reflection of a lack of engagement on the part of Group 2, but were due rather to most of this group contributing to the seminar task outside of the online conference. This was evident from logs of contact and activity which the trainees were asked to keep at the time. These logs consisted of proformas for noting down all relevant activity, who else (if anyone) was involved, the medium used, the date, and the time taken. They therefore provided details on how email communication fitted into other activity and contact during this exercise.

For Group 1, four out of six logs were returned, and these were used to calculate the mean time spent on each of four broad categories of task-related activity:

• face-to-face communication
• email communication (including failed attempts to connect)
telephone communication (including the sending of faxes)
• independent activity (including word-processing and internet access)

This showed that members of Group 1 spent roughly the same amount of time on face-to-face and email communication, and made little use of the telephone. Not surprisingly, a large proportion of their time was taken up with independent activity such as accessing information and preparing draft sections of the seminar paper.

Face-to-face communication was reported to have taken place more typically at the outset, and at points of negotiation, whilst email was used more for the coordination of work within previously negotiated frameworks, the exchange of drafts, updating others on progress, and deciding on the format of the seminar presentation.

This data confirms the other indications that although a reasonable proportion of time was spent conferencing (or attempting to), email was used primarily for information exchange, whilst discussion was reserved for face-to-face contact.

For Group 2, only two out of the six logs were returned. If these are taken as representative, though, they reveal that whilst the total time spent on the seminar task was similar to Group 1, proportionately much less time was spent on email, and more on face-to-face (and fax) communication, consonant with the message frequency data. Face-to-face communication for Group 2 included working together, on top of the strategic uses made by Group 1. Thus lack of email contact via the shared conference did not signify lack of engagement with the task, but a use of other methods of working and other channels of communication.

Experiences during the seminar task seemingly crystallised patterns of usage of the conferencing system in terms of both the function and individual frequency of contributions. This is apparent in the rest of the message archive, both from the initial monitoring period and subsequently, from Autumn 1998 to Summer 1999. Seminar-related messages were of course only prevalent whilst that task was ongoing. In the absence of any other activity with as clear-cut a shared objective as this, message frequency in the conferences never again reached the peak it achieved during this exercise (the average over the period from Autumn 1998 to Summer 1999 was around four messages per week). At the same time, though, the nature of the messages that were sent remained largely the same. Thus the conferencing system was used for other purposes, including group project work and preparation of the conference poster, social messages, the seeking of advice from tutors and peers, and course administration messages. Postings of these types tended to occur with moderate frequency throughout, but in all these instances, it was still information exchange of the kind that had taken place during the seminar task that appeared to be the primary objective.

Interviews with the trainees confirmed that this kind of information exchange was seen as the basic mode of use of the system.

A further indication of this is that, despite efforts to set up a discussion conference (NB instigated by one of the frequent Group 1 users), contributions to this were sporadic (one per week on average), and notably less than those to the general conference, where messages of the type noted above continued at the rate of three per week.

One other striking illustration of the influence of the seminar task was the fact that there was a substantial correspondence between individuals' early and later frequency of contributions. In other words, those who sent most messages during the period of the seminar task itself (primarily the members of Group 1) were also those who subsequently sent most messages to the general and discussion conferences. It was these individuals who in addition reportedly explored use of the chat facility (though apparently only to a limited extent) and other email systems.

It should be noted that throughout the course there was some usage of private mailbox facilities, and that the extent and function of the messages sent to these is not clear, since they were not monitored directly. However, the main distinction drawn between contributions to the shared conferences and private messages seems to have been simply one of whether or not the content was personal (in the sense of targeted rather than sensitive), which would suggest that the pattern of communication was unlikely to have been very different.

Self-Perceptions

As with the affordances and constraints imposed by the system, there was little explicit comment by the trainees on their perceptions of themselves as contributors in relation to each other, or on how this might have affected the messages they sent to the conferences. This may be a reflection of a combination of factors, including the relatively small size of the group, the fact that they knew each other well, the closed nature of the system, and the emphasis on informational exchanges which required no real statement of personal positions. Certainly there were no accounts of confusions, misunderstandings or conflicts which would have heightened self-consciousness, and what comment there was focused solely on the general principle that making any communication necessitated thinking about content and audience)

A further factor here may have been that even the output from the seminar task was not formally assessed, so that at no point were there any externally derived reasons for sensitivity about the quality of contributions.

Of course, one consequence of this may also have been that the trainees felt no particular pressure to make contributions either, other than to satisfy their own perceived needs, and this may have had some bearing on the relatively low level use of the system.

Learning and Teaching Relationships

The lack of self-consciousness did not, however, extend to awareness of the trainee cohort as a whole, and how bonds between people coloured usage of the system (and presumably vice versa, though this was less explicit). Since so much of the online interaction focused on information exchange, not surprisingly the key concern was the rate of response to requests for information, and the time lag between request and response. In general, the cohort was seen as being a good resource in this respect.

At the same time, however, there was plainly some differentiation between individuals in the extent of their responses, in part because of access problems, but also perhaps because of the attitudes and habits that the trainees had adopted with regard to system use.

One consequence of this was the emergence of a sub-group of trainees (NB mostly consisting of members of Group 1) who made more effort to respond to each other, the resulting positive feedback then helping to maintain patterns of use.

This is not to imply the emergence of a clique, however. Those who were part of this sub-group were as careful to respond to requests for information from those outside it as they were with each other. One of the course tutors also played an important role in encouraging responses too.

There was also some effort to provide positive justifications for a lack of response from some people, and to downplay the negative consequences.

Nevertheless, those who became more regular users of the conferencing system did help sustain each other, and it is hard to imagine that this had no impact at all on dynamics within the cohort or on the resources that were available to different individuals.

Getting Things Done

A further element feeding into trainees' use of the conferencing system was their prior experience of task-based communication. In order to gain a picture of the trainees' communicative styles and capabilities under more standard circumstances, observation of a face-to-face seminar was undertaken prior to the setting up of the system. The seminar chosen was the last prepared without potential email support, or allocation to task groups. This observation illustrated both the trainees' familiarity with each other, and the fact that they were experienced face-to-face communicators with similar ways of approaching the presentation of information. Over a two hour period there were lengthy contributions and substantial interjections made by almost all present, with the trainees addressing both contributions and interjections to each other rather than to the tutor who was chairing the session, as would typically be the case amongst less experienced students. Also striking was the incidence of spontaneously organised floor-shifting, without direction from the tutor: those who had not yet contributed would typically pick up on a point from the previous speaker, develop it and carry it on into their own mini-presentation.

Less in the way of direct information was available on the trainees' prior familiarity with joint writing exercises of the kind which the seminar task required them to undertake. However, they were all graduate students in a discipline which places considerable emphasis on writing skills, and they had been set other joint writing tasks earlier on in the course. Thus a certain degree of expertise and similarity of approach can be assumed.

Any facility the trainees had with face-to-face communication and joint writing was at first sight of limited assistance when it came to engaging with online communication, however. This point was underlined by responses to a brief questionnaire each group was asked to complete after they had finished the set seminar task. This dealt with:

• the trainees' experience of email prior to the course
• their initial confidence about carrying out the seminar task
• the factors affecting this, and what additional support they felt might have been useful
• the perceived usability of the conferencing system and software
• its main advantages and disadvantages
• which aspects of the seminar task had been seen as most useful, and which as least
• whether they had modified their approach to the task because of the online element
• whether there were any surprising or disappointing outcomes
• what factors they saw as determining the success of a conferencing system?

Taking Group 1 and Group 2 responses separately in view of the other identified differences between them, for Group 1 only one of the five respondents reported having utilised email previously. Despite this lack of past experience, on balance the members of this group said they were moderately confident about the seminar task at the outset. However, most also said they had had doubts about the software and its reliability, and all respondents stated that more initial instruction in First Class (especially sending attachments) would have been helpful. For Group 2, two out of the five respondents reported having had previous experience with email (so NB the differences between Group 1 and Group 2 were not attributable to the former containing more experienced users). Again on balance the group members had been moderately confident about the seminar task at the outset, but in general the pattern of other responses was similar to those given by Group 1.

This theme of inadequate training was subsequently picked up in interviews with the trainees, where it was characterised as being in part a reflection of the fact that the conferencing system had been a late introduction to the course.

The perceived inadequacy of the training was also blamed to some extent on the course tutor.

Yet despite these perceptions that greater training and instruction would have been helpful, it was in fact the case that when there was a need to send attachments, cut and paste text, or access other conferences, Group 1 (but less often members of Group 2) worked out how to do so, simply by exploring use of the system. They then let each other know what was involved. In other words, then, familiarity with the communications required by a task was a benefit, because it led to an awareness of the actions that had to be carried out to move that task forward, and a deliberate effort on the part of at least some of the trainees to establish ways of doing so using the system.

There was even some feeling that more formal training might in fact have its limitations.

It is important to note that this process of exchange and building up of group expertise was perhaps critical to the success of this task-driven exploratory approach to getting on top of the software. Certainly, in the absence of this kind of exchange, trainees appeared to feel isolated and unsure of how to proceed.

Of course, what kind of tasks the trainees considered using the system for and how these were executed may itself have been subject to external influence. As a check on how its usage was initially framed for and perceived by the trainees, written records were kept of the early meetings where the introduction of the system was discussed, and of the trainees' reactions during these. These records were either notes of direct observations made at the time, or notes of interviews with course tutors immediately afterwards.

The idea of using email and conferencing facilities was first mooted by the course director at the initial email demonstration, prior to trainees commencing their placements. Possible functions it could serve which were highlighted at the time are noted above (see 'Stated purpose of the conferencing'), and these apparently had some impact, since they were echoed during the interview sessions more than a year later.

At the same time, though, these external influences have to be set alongside the effects of the trainees' own explorations. Once the conferencing system was set up, the course tutors left an interval for the trainees to investigate its operation (after they had received a basic training), before setting specific tasks that might require its use. The intention here was to allow them to establish for themselves something of what they might be able to do with it.

Similarly, when the seminar task was introduced, whilst the groups were provided with a clear objective (and NB one which was familiar to them from past work), the manner in which this objective was pursued was left to the groups themselves to determine, beyond a recommendation to use the conferencing system as part of the exercise.

Thus, on the one hand, the pattern of usage of the conferencing system that eventually emerged, especially around the seminar task, makes it tempting to connect this back to the fact that support for joint writing was one of the potential functions highlighted during the introductory sessions, whereas debate and discussion was not particularly mentioned. On the other, it is possible that this had no influence. It may simply have been that because the group had the opportunity for face-to-face contact throughout, it seemed natural to reserve discussion for these occasions, and to use email for the textual exchange/commentary and informational exchanges to which it was well-suited. This fit between the channel and the activity would have been especially high with respect to the seminar task itself, which had the specified objective of joint construction of a text.

Moreover, it is likely that the trainees were already aware from past experience of the need for exchange of sections/versions of texts as part of progress towards this objective. Overall, then it seems safest to conclude that the perceived functions of the system were shaped by more or less explicit external direction as to appropriate tasks, together with trainees' own understanding of those tasks and how to go about them. However, it does also seem to be the case that any fluidity there might have been in setting up the system functions was relatively short-lived. As noted above, once the informational exchange function of the system had been established, particularly as part of the seminar task, it appeared to be sustained during subsequent activity.

This continuity was illustrated by, for instance, the paucity of input to the discussion conference (the existence of which some trainees were even unaware of), and the fact that what contributions there were typically took much the same form as contributions to the main conference, although there were some attempts to open up more discursive input. Textual exchange per se was rarer, but there was less obvious requirement for it in the second year of the course.

One further point to note is that the nature of the information exchange 'task', and of the function served by the system as part of this, was such that it required almost no explicit coordination, and could rest simply on sufficient numbers of people logging on frequently enough to maintain a stream of requests and responses (see 'Learning and Teaching Relationships'). Where more explicit coordination was necessary, within the seminar task, decisions about basic strategy and division of work were negotiated face-to-face, and activity was coordinated online in a finer-grained way via reference to these decisions (see 'Character of the Communications' above).

In short, then, once tasks or purposes had been negotiated or tacitly understood, these provided shared frameworks for the implicit organisation of exchanges, obviating the need for explicit coordination of activity via online messages. It seems likely that this process was facilitated by the task grounding that the trainees had in common from the outset.

Continuity with Curricula

As noted previously, the conferencing system was seen by the trainees as something of an add-on to the rest of the course, and thus as standing at some remove from its direct educational content and objectives (though see 'Evaluations' for a more implicit objective that it did relate to well).

This was true even of the seminar task, which was recognised as being merely an attempt to provide a reason for using the system.

This is not to say that there was no awareness of how the system might be utilised for more formal course support, but there was frank admission that how it had been used was some way away from this. Again, this is something that might have contributed to the relatively low level of use of the system.

Evaluation

All the Group 1 respondents to the post-seminar task questionnaire said they liked the speed of exchange which email allowed, and most also mentioned the convenience of asynchronous exchange, which meant not having to wait till someone was in, and having time to think about replies. Comments during subsequent interviews reiterated these points.

The successful joint production of the seminar paper was also spontaneously mentioned in the questionnaire by almost all of Group 1 as a surprising (and positive) outcome from use of the system. In contrast, Group 2 made no mention in their questionnaire responses of the benefits of asynchronicity (perhaps reflecting paucity of use of the system), and had a less positive reaction to the resource as a whole, not least because of the impact on usage of the initial problems with the setting up of the system.

These general evaluations were an accurate reflection of the problems that all had had with the operation of the system, but also of individuals' different ways of coping with these. Thus, throughout the initial five-month period of monitoring of usage of the conferencing system there was one category of message in addition to the others (see 'Character of the Communications' above) which had a consistently strong presence. This category was exchanges about network connections or software difficulties, either pointing up problems and seeking advice; or else announcing the solution of problems and the establishing of connections. During 11 out of the 17 weeks from the start-up of the system until the summer vacation this was the most frequent or joint most frequent category of message. In other words, then, all other activity during this time took place against a background of perceived connection problems, though individuals varied in their reaction to these.

In line with this, all the Group 1 respondents to the post-seminar task questionnaire reported having experienced access problems due to network difficulties or lack of availability of machines, and noted that they considered dealing with these problems as key requirements for the success of such systems. Group 2 placed even greater emphasis than Group 1 on technical problems, and reportedly split into pairs to carry out the seminar task in order to get around them, hence the greater incidence of working together noted in the logs, and the lower use of email.

In part, these reactions were a genuine reflection of the start-up difficulties noted in the Background section above. However, the perception that there were problems using the system persisted beyond the point at which connections from placement offices had been reliably established.

This was particularly true for Group 2, which contained the person who had the greatest difficulty establishing connection at the outset. Despite the fact that these problems were solved by the time the Group 2 seminar task was set, they still evinced a greater reluctance to make use of the conferencing system - and indeed apparently still had doubts about its reliability months later.

Given the functions of email for Group 1, one factor which might have contributed to this longer-term effect (and thus to the broader differences between the groups) was that Group 2 reported more failures in sending and receiving attachments. These failures were to some extent due to mismatches between the resident word processing software (Word and ClarisWorks) on the group members' machines. They also appeared, however, to reflect some lack of understanding of how to circumvent these difficulties (e.g. by standardising software, or by using rtf files as the exchange format).

There were also reports of continuing problems with access to machines, especially in the placement offices, where demand for resources meant that they were frequently tied up in ways that meant linking to the conferencing system was not possible. However, there were also some signs amongst these reports that, perhaps because of initial negative attitudes, some people were not trying that hard to make connections.

Whatever their precise nature, the implication is that technical difficulties led Group 2 to adopt a structure for the seminar task which minimised the need for exchange, and that this was the primary reason for the differences in interaction patterns between Group 1 and Group 2, both at the time and subsequently. If so, it is not clear whether these technical difficulties were more perceived or actual. One possibility is that Group 2 had greater lack of confidence in the system because they had fewer early frequent users: three in Group 1, against one, or at most two in Group 2. As a result, Group 1 may have had a "critical mass" of users which was sufficient to persevere and overcome any lack of confidence in the system, whereas Group 2 did not. In a sense, though, whether this account is accurate matters less than the fact that, whatever the cause, due to technical problems Group 2 negotiated a working procedure which left the conferencing system without any clear function. As a consequence, in contrast to Group 1, they used it much less, and this became an established pattern.

As noted above (see 'Continuity with Curricula'), the system was not seen as addressing in any direct fashion the primary educational objectives of the course. However, there was one more secondary objective that it was seen as fulfilling, namely the more generic concern of learning how to handle such technologies in preparation for professional employment. Thus responses to the post-seminar task questionnaire indicated that members of Group 1 all felt that learning about conferencing had been the most useful aspect of the seminar task, and that keeping in touch or discussing were the least useful aspects (cf. the relative lack of use of the conference for this, and the availability of face-to-face contact). Group 2 at that time perceived no real advantages to having the system, but subsequently there was a more general agreement that having had experience of conferencing might be useful for helping to establish the kind of information exchange resources available to other groups of professionals.

Moreover, for some of the trainees there was the additional motivation of not only knowing how to use such systems, but also having the edge of being amongst the first in Service offices to have such knowledge, something that was unlikely to be a barrier to subsequent promotion.

However, the comparison to the existing activity of professional educational psychologists may have been a double-edged sword, something that was actually demotivating for some of the trainees, because they saw the potential redundancy of a skill that was not generally valued within the culture of which they were shortly to become an established part.

Thus whether the system was seen as a salient and useful part of the course provision may have been influenced by calculations about the likelihood of the culture changing, and the possibility of contributing to that process of change. This too may have had a bearing on the extent to which individual trainees made use of the system.

Conclusions

This case reveals the complex interplay of factors determining whether text-based conferencing resources are used, and whether the communications that take place are effective and productive. Across the period during which the system was available, the rate of contribution per individual to the shared conferences was patchy, and was not high for any. However, the trainees' workload meant use of the conferencing system was always likely to be restricted, and since they had face-to-face contact two days per week, it was often not necessary at all. The effect on message frequency of the seminar task demonstrates that they did make use of the system when there was reason to do so, underscoring the particular influence of having clear objectives and tasks on the incidence and effectiveness of online communication. In the absence of these - and few other tasks had such clear objectives - use was fairly minimal.

This influence may have been accentuated to some extent by the trainees' initial reluctance about the introduction of the conferencing system, which may have led some of them to feel especially disinclined to explore its usage without a very good reason to do so. The lack of assessment of online activity or its products, and its apparent lack of value to the wider professional community may have further led some to conclude that such reasons did not exist.

Individual differences in prior experience of text-based conferencing, on the other hand, seemed to have no direct impact on the trainees' pattern of system use. Moreover, despite perceptions that greater training and instruction would have been helpful, in the absence of this, when trainees themselves (especially Group 1) wanted to achieve something using the system, they typically made it happen.

Overriding all other influences, though, was the perceived need for the system, which was affected in turn by its characteristics in terms of ease of access and reliability. In the absence of external incentives, usage of the system gravitated towards those functions and needs it could apparently serve, such perceptions being coloured by external suggestion, but more particularly by previous task and system experience. Thus it was not so much that the seminar task actually fixed the emphasis on information exchange, but rather that this was one need the trainees knew they had, and the seminar task had provided them with the knowledge of how they could use the system to meet it.

Whether they saw it as a worthwhile exercise to use the system in this way, however, was strongly influenced by perceptions of its accessibility and reliability. All the trainees had problems on both counts, but Group 1 saw a useful role for conferencing in spite of these, especially as part of the seminar task, and so fell into a habit of employing it. Group 2 apparently mistrusted it more, declined to give it a role, and so used it very little, either then or later. It is important to note that the association of differences in usage of the system with group membership and with the scale of perceived technical problems is a clear indication that these differences were not simply a reflection of individual variation in initial intentions to engage with the system.

What remains less clear is quite why the two groups differed so markedly in the working procedures they adopted for the seminar task, given that they had the same goal and similar backgrounds and experiences. It is tempting to point to the initial suggestion by the course tutor that the conference system could be used to support joint writing as having had a bearing on later decisions, in view of the way in which it was eventually employed. However, differences between the groups in the take-up of that possibility appear to rule this out, and in any case the tutor's suggestion was not accompanied by discussion of how this might work in practice.

There is a tacit assumption here, though, that the groups made firm, clear choices to adopt their different procedures, and this may be overly simplistic. One sign that it was not the case is the fact that Group 2 did make tentative efforts to use the system in the same way as Group 1, but relied primarily on other procedures (perhaps those they had used previously to produce joint presentations). Prompted by the tutor's suggestion, then, the system may have appeared to them also as an attractive means of achieving the task objective, as long as it could be trusted to be reliable.

Similarly, for Group 1 the renegotiation points and their surprise at achieving the goal in this way both suggest that they did not in fact start out with a clear model of how to use the system effectively for joint writing. Instead, it seems likely that they too had ideas about how joint writing should be conducted based on past experience, and, pursuing the tutor's suggestion, worked out how to adapt these to the circumstance of having the conferencing system available. They may have been aided in this by the particular suitability of the system (in technical terms, the "affordances" it offered) for the exchange of text.

If this account is accurate, it suggests that Group 1 exhibited an agreed transformation of working practices which Group 2 failed to achieve (though their approach was consensual too), because it depended on the perceived reliability of the system, as well as on its possible uses. Thus the division between the groups may have come down in the end to a marginal difference in just this respect. Both saw the system as unreliable, but Group 1 found it trustworthy enough to persevere, whereas Group 2 did not, and the consequences snowballed on from this branching of perceptions.

It is worth pointing up the apparent ease with which the perceived instability of the system deflected Group 2 "off-course", where more seasoned users might have been less affected. Although lack of prior experience with email did not appear to be of much consequence in itself, it may be that the interaction of this with small differences in perceived reliability affected the outcome in this case. New users are perhaps in this sense unstable themselves: any influence, especially a negative one, has a larger impact. Moreover, once this initial deflection had occurred, it was apparently very hard to make up lost ground, since 15 months later the differences in the extent of usage of the system were still present. This suggests that initial problems led to deep-seated distrust of the system, or at least a habitual passing over of what it might have to offer.

There are a number of pointers for the success of future implementations flagged up by this. Firstly, there is a paramount need to design resources where from early on students can readily establish clear and meaningful joint objectives that encompass use of online conferencing, especially if there are no other strong motivators to engage with it. As part of this, it may be useful to ensure that such resources are available from day one, so that students do not get used to one pattern of working, only to be expected to alter this with the advent of online facilities. It is also important to bear in mind that promoting use of conferencing in this fashion does not mean that online discussion will inevitably be the result of this: other types of productive activity may occur, depending on the objectives arrived at.

Secondly, there is a need to secure the reliability of the system across all points of use at the outset, especially if the students involved are relatively inexperienced, since even the perception of unreliability may have extended negative consequences. Under these conditions it may also be helpful to establish the optimum operating parameters of the system, including e.g. the format for attachments; and to ensure that information concerning these is passed on to students.

Finally, for both these points, the importance of initial activity is emphasised, since the indications from this case are that, especially with new users, first experiences have a considerable impact.

Text of this section by Andy Tolmie, editor Erica McAteer

©Erica McAteer, Charles Crook, Andy Tolmie, Hamish Macleod, Kerry Musselbrook, David Barrowcliff, 1st May 2000