Talking with colleagues during the week about the data we have, brought it home to me how incomplete our view of our electronic resources use actually is.  The various systems that we have to record usage all give us a different perspective but all have limitations.   If we think about a simple requirement, which would be to know who is using our resources, broken down by which resource, from which supplier, and whether they are students or staff, and what course they are studying, then its simply not easy to pull that together in a simple way.

EZProxy logfiles
We use EZProxy wherever we can and the daily logfiles tell us what URL has been accessed and who the user was, when the site was accessed, and where they came from.  But not all our electronic resources go through EZProxy.  Tools such as RAPTOR, officially launched recently, allow EZProxy logfiles to be processed and statistics generated of which databases are being used.   Unfortunately for us RAPTOR is largely designed for a federated search world where electronic resource access largely goes directly to a wide range of publisher websites.  In our case we use a Discovery system (in our case EBSCO Discovery Solution) so most of our article level links show up in the EZProxy logs as links to EBSCO.  So at this stage RAPTOR just tells us that our resources are being delivered from EBSCO.

Athens logfiles
For some of our resources we still use athens and we can get similar data to the EZProxy logfiles, although the data is provided externally and downloaded monthly.

Directly logged-in resources
Other resources can only be accessed through the use of specific user names and passwords.  For these sites all the data will tell us is that the site was accessed, how many times, full text downloads etc, but it can’t tell us who is using the site, as we have to rely on the statistics from publishers and don’t have access to the logfiles.

Publisher statistics and JUSP
Use of electronic resources is also monitored by the various suppliers.  Some of this data is added to the Journal Usage Statistics Portal which saves some time that would have to be spent in collecting data from individual publisher sites by providing a collected Counter-compliant set of journal statistics.  But while tools like this show us the overall use of those resources, they don’t help if you want to drill down to see who is looking at those resources.  Are they students or staff, are they students from this faculty or another?  And they also seem to suffer from some issues around the impact of Discovery systems generating additional queries (leading on some occasions to suppliers shuting down access because they think they have detected unusual patterns of usage).

Other data
Links to library resources are also present and tracked in the institution’s Virtual Learning Environment.  In this case VLE logfiles might tell us who has accessed the resource but the details of which resource has been used is limited to the URL and it apparently isn’t possible to track every single location within the VLE where we place links.

An incomplete view of the world
So what might help us?   Based on experiences with the RISE projct with the EZProxy logfiles we can use some techniques to add bibliographic metadata to the EZProxy URLs.  That can give us article details, Journal titles and publisher information, for example.   Combine that with the data we can get about the user to pick up their course or faculty and that gives some potential to be able to answer our original question  “who is using our resources, broken down by which resource, from which supplier, and whether they are students or staff, and what course they are studying”.