HOME
Results
It had previously been decided, from looking at the available literature, that eight weeks was an adequate period to give accurate information concerning numbers of users, identification of users, and possible trends in user behaviour, and that the first eight weeks of availability was the most suitable choice for the obtaining of useful information. Accordingly, on February 9 an announcement concerning the online availability of the abstracts was made to all participants, and also on relevant online mailing lists.
Ten weeks later, a breakdown of log-ins (Fig. 1) from the selected eight-week period was obtained. From the breakdown, it was not always possible to identify the user, nor even the country of origin because some email addresses bear no features which the software in use can identify. Nevertheless, since the overall results are interesting, and their impact does not depend on identifying individual users of countries, it was felt to be worthwhile to communicate these to other libraries at this point in European development of online library resources.
The information obtained also identified which part or
parts of the MTP workshop had been accessed by the users,
and how much time they had spent browsing at each stage.
Using this information it was possible to analyse the data
to investigate how and when the abstracts were being
exploited by the users, and to identify which projects had
generated most interest. Because of a technical problem it
was not possible to obtain full data from March 20 - March
26 inclusive, and therefore the data for that week have not
been included in the survey.
Figure 1 shows that there was high initial
interest, almost four hundred accesses in the first week,
then 115 in the second followed by two fairly similar
figures in the 90s, then a drop to 50, then even fewer
accesses, followed by another slight rise ( this latter
accounted for by the fact that the Opening Remarks of the
MAST Team was put online during that week). Figure 1
also indicates that both the number of hits and the number
of users dropped consistently over the first seven weeks of
the study, with only a slight increase noted for the eighth
week.Throughout the study period there were fewer accesses
at weekends, particularly on
Sundays.
This type of measurement, though it is the type normally
used as a criterion of online success or failure, is
nevertheless rather misleading, as can be seen from
Figure 2. There the number of actual users of the
abstracts shows a very noticeable difference. Those who used
the abstracts are now seen as comprising a mere 62 out of
the initial 400, and this drops to a steady, and declining,
weekly rate around the twenties. In this context, however,
it must be noted that there were in total less than 150
participants in the MTP workshop. We also recorded
individual abstract users, of which there were 175 who made
a total of 856 accesses. Where the addresses could be
identified to individuals, we found a number of MTP workshop
participants, but we also found many
non-participants.
Where users can be identified to country (Figure 3) we also found an interesting situation.
The first left-hand column shows that from the
identifiable users, North America had the highest
representation despite the fact that there was only one
participant from that region. The Mediterranean countries,
and other participants, were reasonably well represented,
with quite a number of Greek and British users, and also a
significant number from France, Spain and
Portugal.
The abstracts consulted were then identified (Figure
4). Most accesses, almost 200, were made in order to
consult the index. It could be argued that these were users
who were not participants and therefore needed to consult
the index. The next highest amount of accesses were made to
the project MERMAIDS, known to be one of the most successful
of the MTP projects. The next highest number of accesses
involved the IMBC project, CINCS. Opening remarks and
graphics were not added to the site until the fourth week of
the study, so figures for these features are not entirely
relevant.
From Figure 5 it is possible to ascertain how much time users spent browsing through the workshop link, thereby indicating the degree of interest generated by the abstracts, and their usefulness to the users. While the majority of users spent under two minutes in total, or merely accessed the index, there was also a substantial number who spent a considerable amount of time looking at more details.
Introduction | Methods | Results | Discussion
WHAT'S NEW - THE INSTITUTE - PROJECTS AND PROGRAMMES - CASE STUDIES
BIBLIOGRAPHICAL SERVICES - PRODUCTS & SERVICES - INTERNET RESOURCES