I'm going to take a controversial and extreme point of view here.
You should not pay for statistics from your vendors because the
statistics are worth ABSOLUTELY NOTHING. Nil. Nada. Zip. Actually,
they are worse less than nothing, because they are sure to lead you
to the wrong answer to almost ANY question. (Having said that, I must
admit that I'd want to see the data, anyway, just like you)
Consider the simultaneous user question. suppose the vendor reports
that your number of simultaneous users has doubled from one months to
the next. What does this mean? It could mean that usage has doubled.
It could also mean that the session timeout has changed, or it could
mean that the vendor's system was overloaded, causing users to spend
twice the amount of time at the site. You don't know. And you might
not be surprised to find out that that your vendors often don't have
a clue either.
When I first started running an e-journal, I watched the usage logs
compulsively. "Oh look, someone from Yemen!" I'd exclaim. One day I
noticed that hits were running 3 times the usual. "Finally" I
thought, "the materials research community is getting excited about
our e-journal." So, being a scientist, and an experimentalist at
that, I did some investigation. I turned on referrer log, and found,
to my chagrin, that people visiting the site had been asking search
engines for information about "Lucent", the company where I worked.
We had just been split off from AT&T, and it turned out that our
e-journal was one of the few sites in the search engines that
mentioned Lucent. 0.1% of 10 million people was overwhelming 20% of
5,000.
The scientific method is appropriate to this situation. Pose a
question. Design a measurement to answer the question. Use controls.
Eliminate variables.
The best methodology to answering the question "which of my resources
is used the most?" is not asking the vendors for statistics. In the
best of circumstances, the data will be apples and oranges, and more
likely the data will be cows and giraffes. The very best method is to
analyze logs of the proxy server for an intranet.
Jonathan Eaton gives some excellent advice. You can also use a
rewriting proxy server like EZProxy as a usage logger. Services which
have implemented OpenURL awareness give you another way to
"recapture" usage data.
Eric
At 9:09 AM +0000 3/15/02, Maureen Richardson wrote:
>Could I have your views on this? As many of you will be doing, I am
>preparing my budget proposals for next year and again as I am sure many of
>you are doing, I am looking at usage statistics of journals and online
>databases. I routinely email the suppliers of resources who do not provide
>me with the facility of retrieving my own statistics. I contacted one
>supplier who told me they would charge 50 pounds for this information and 50
>pounds each time I requested it! I was stunned. The ironic thing is that we
>have no doubt that this is a well used database and what I actually wanted
>was the number of simultaneous users and turnaways. We were in fact
>wondering if we should increase the licence. We pay 1700 pounds for this
>database and I think we should be entitled to statistics. Have colleagues
>ever had this reply before and what has been their response? It is an
>essential database and we cannot cancel. There is another supplier but they
>cost more.
>
>Thanks
>
>Maureen
>
>Maureen Richardson
>Journals and Electronic Resources Manager
>Learning Resource Centre
>Edge Hill
>St Helens Rd
>Ormskirk
>Lancashire L39 4QP
--
Eric Hellman, President Openly Informatics, Inc.
[log in to unmask] 2 Broad St., 2nd Floor
tel 1-973-509-7800 fax 1-734-468-6216 Bloomfield, NJ 07003
http://www.openly.com/1cate/ 1 Click Access To Everything
http://my.linkbaton.com/ Links that Learn
|