RE: [IUG] questions about creating statistics reports in millennium

["Seeman, Corey" <cseeman at bus dot umich dot edu>]

Joe (et al) -- 
 
The Milstats program is the updated version of the character based function called "Create Statistical Reports."
 
The basic function of these reports really has not changed (although the functionality is vastly improved over character-based).  These reports really are intended to be a place where you can take a large group of records (from a range, review file...or in Millennium from an index search) and get record counts.  It is much cooler than that, but it serves as the quick description.  Since these reports live in the present, they can be run and run again.  The nice thing about Milstats is that old reports stay in your system...something that required you to print reports from the character-based version.
 
The circulation data (raw numbers) will be kept for 3 years for statistical purposes.  I am not sure what data is being missed by the system when mention the "lack of an extensive historical data file for Millennium Statistics".  Some of this work can be pulled out of a system (when a record was created, and by any of the date-related fixed fields such as CAT DATE, RDATE, etc.)
 
As a newcomer to the Innovative system, there are certain expectations that you might have had with your old system.  These might make for good enhancements to the Innovative System.  The process that is sponsored by the IUG for enhancements is an excellent one that puts the users in the driver's seat.  We propose these enhancements; IUG members serve on the committees that review them and then we (as a membership) vote on them.  It gives the company excellent information about what is really important to the users.  The preliminary enhancements documents released by Innovative state which ones came from the IUG process and that is nice for us because it lets us know that the process is working.
 
So if there are areas where it is not recording information that is critical, by all means submit them as enhancements.  The URL is http://www.innovativeusers.org/enhancements <http://www.innovativeusers.org/enhancements> .  
 
The new eyes on the system sometimes see things that we are not looking for!
 
Best -- Corey
 
 
Corey Seeman
Associate Director for Information Resources and Systems
Kresge Business Administration Library
Ross School of Business
The University of Michigan
cseeman at umich dot edu
(734) 764-9969
 
Kresge Library Home Page: http://www.bus.umich.edu/kresgelibrary/ <http://www.bus.umich.edu/kresgelibrary/> 
My home page: http://www-personal.umich.edu/~cseeman/index.html <http://www-personal.umich.edu/~cseeman/index.html> 
 
 
 

________________________________

From: innopac-bounces at innopacusers dot org on behalf of Hewitt, Joseph W
Sent: Mon 1/30/2006 12:24 PM
To: IUG INNOPAC List
Subject: RE: [IUG] questions about creating statistics reports in millennium



Corey,

Thank you for generously taking time to explain this Innovative
statistics thing to me. On Friday, I just wasn't getting it.

This morning with a clearer head I studied your posting and reran my
reports. Our installation of the Web Management Reports feature does
show a separate column for renewals and the checkout total matches my
Innopac result which makes me happier. So, I suppose, point one is moot.

As for the Millennium transaction file, won't the lack of an extensive
historical data file for Millennium Statistics continue to worry library
managers as more features move to Millennium?

Are you saying that MilStats has no audit trail of database activity?
This omission would not be good. As you undoubtedly know, administrators
are often asked to take a long view of the data for trend analysis and
factual arguments for university administrators or library boards. Lack
of an audit trail would require local libraries to create in-house,
idiosyncratic back office tracking systems, especially if they sense a
loss of access to the Innopac history file. (Do I understand this point
correctly?)

Thanks for confirming my skepticism about the usefulness of stat groups.

Your note cleared up several poorly-documented points. I think I'll go
look at your presentation now.

Thanks for your patience as I try learn more about Millennium,
Joe


Joe Hewitt
Library Systems Specialist
Missouri State University
901 S. National
Springfield, MO 65897
417 836-4536

-----Original Message-----
From: Seeman, Corey [mailto:cseeman at bus dot umich dot edu]
Sent: Saturday, January 28, 2006 6:30 AM
To: innopac at innopacusers dot org
Cc: Seeman, Corey
Subject: RE: [IUG] questions about creating statistics reports in
millennium

Hi Joe --



Welcome to the Innovative family.  I think I know what might be going on
and that might explain what is happening with the numbers.  In looking
at your query, I checked the numbers at my library and my former library
and found that they were looking different.  A bit of digging and a call
to some friends who really know things cleared it up.  I will try to
keep everything together, but will refer to your original query below.



I would like to write about to main issues:



--The discrepancy between telnet (character based) and Web Management
Reports).

--The difference between Circulation Statistics and Millennium
Statistics.



First, about the discrepancy in the telnet (character based) vs. Web
Management Reports.



As they are designed, the Web Management Reports are the next generation
(or Millennium version of) Circulation Statistics.  They are pulling
from the same data on the Innovative system, but maybe doing things a
bit differently.  Hopefully, someday they will move to the Millennium
client and we can easily export them like we can with Milstats.  So I
want to talk about this first.



This is one of those situations where I had no idea that this was
potentially a problem!  So it is great when we get to learn something
new about the system.  For my old system, I was looking at Circ Activity
(All) with telnet and Web Management Reports and the totals appear to be
sync between the two versions.  However, when looking at All Activity in
telnet does include renewals as a unique column, whereas in the Web, it
does not show up.  What is happening is that the renewals are being
added to the checkouts.  So for me, the totals are fine, but the
Checkout totals are totally off because one includes renewals and one
does not.  I am not sure, but this could be where you are seeing a lack
of consistency between the numbers. 



This might be where the problems in your reports are coming from.  Here
is one example:



Telnet:



CIRCULATION (CHECKOUT) STATISTICS

--- Jan 27,06 -----

P TYPE                       

Graduate Student          114

Undergraduate               102



CIRCULATION (RENEWAL) STATISTICS

 ----- Jan 27,06 -----

P TYPE     

Graduate Student          19

Undergraduate              13



But via the Web Management Reports -- You see these combined under
Checkouts (sorry, I did not say there would be math on this email).



CIRCULATION (CHECKOUT) STATISTICS

Jan 27 06

P TYPE PERCENT QTY

Graduate Student 30.4% 133

Undergraduate 26.3% 115



Libraries can have three different ways of viewing these statistics via
Web Management Reports.  This can be setup via the helpdesk's Service
Commitments.



1. Renewals separate from Checkouts (Circ Activity and Circ Crosstab)

2. Renewals combined with Checkouts (Circ Activity and Circ Crosstab)

3. Renewals combined with Checkouts, in All Activity report "Items Circ"
column (Circ Activity only)



You can go to the Service Commitments and go right to the bottom
(Statistics). 

http://csdirect.iii.com/calls/services/commit.php?mod=3&name=Circulation

There is an option for: "Combine or separate checkout and renewal
statistics for Circulation Activity and Cross Tabulation Report."
Definitely talk with the helpdesk about this if you (or anyone is
confused about these options).  I already suggested that this section in
the users manual be updated to reflect the differences between the two
flavors of stats. 



If there are problems with the stats that go beyond this, there could be
other factors at play including INN-Reach transactions or a real problem
with the way that your system is working.  When statistics are not in
sync, give as much detail as you can and open a call with the helpdesk.




Second, the difference between Circulation Statistics and Millennium
Statistics.



This is a big issue but these numbers are really never going to be in
sync, but there is a good reason why.  The nature of the way that these
statistics are generated is very different.



With Circulation Stats (either via telnet or web reports), these are
transaction-based statistics that are recorded when the checkout, check
in, hold, etc. takes place.  The system stores 37 months worth of stats
(so you can have three full years plus the month you are on).  All of
these stats are based only on the coding that is on the system at the
time when the transaction takes place.  This is very important because
we do (and should) change codes from time to time.  If you change codes
in the fixed fields, SCAT Tables, or login stat numbers, the system does
not go back and rewrite the circ transactions based on the new
information.  When I was a trainer, we called this "invalidating the
data".  However, you did this if you needed new codes or a new SCAT
table, but you needed to remember this as you moved forward.  The
transaction based stats record the codes for the fixed fields and the
SCAT category, so it cannot translate as the new entries are added,
changed or deleted (referred to as "bad code"). 



With Millennium Statistics, you have record-based statistics that are
generated when the report is run.  Milstats is really looking only at
the data that is in the system now and is only partly useful for circ
analysis because it cannot show when the circ transactions were made.
So there are a number of instances where these two reports might not
match up.  For example, you could have deleted items that circulated.
Those items, if they circulated in the last three years, will be part of
the circ stats, however, they will not show up when running Milstats
reports.  Another example is if you changed codes or fixed fields.
Milstats will only give you information based on the current codes when
the report is run.  For Acq people who are still reading (hi), this is
also why Milstats numbers do not match with payments if you use Adjust
Fund Balances to tweak your accounts over the year -- no records to
re-run the searches on. 



For an INN-Reach site, this is an important distinction to make between
the two reports.  The items that you circulate from INN-Reach are not in
your system (they will show up as a virtual item attached to the patron
record only while the request is active).  They will show up as a circ
transaction, but will never show up in Milstats because they are not
regular records in your system.  Depending on how many INN-Reach
transactions you have a year, this could make for quite a big difference
in your stats.  Same would be true for ILL transactions. 



One last thing while on this topic.  Stat groups will be of little use
in Milstats because they are tagged to the circ transaction based on
where it took place.  This is set under the logins and has nothing to do
with the records themselves.  So Milstats will have no way of
understanding or caring about where transactions take place.  If you
have multiple locations and the items circulate freely at both locations
(and holds move back and forth, etc.), the location information in an
item record will not correspond to the stat group number that is based
on the login.



I do have a presentation that provides a nice overview of Stats from
Innovative that I gave last fall:

http://www-personal.umich.edu/~cseeman/NOTSL_2005_Statistics.pdf



I am still working on this subject and hope to have some nicer resources
down the road.  I am hoping to work on (in this order): web-resource;
article; novella; screenplay; novelization of the movie; and then
Broadway Musical (Nathan Lane is very excited, but Mathew Broderick is
definitely lukewarm.  But that is OK, the restraining order expires in
three years so that gives me time to work out the kinks).



Hope this helps in the meanwhile.



Best -- Corey

Corey Seeman
Associate Director for Information Resources and Systems
Kresge Business Administration Library
Ross School of Business
The University of Michigan
cseeman at umich dot edu
(734) 764-9969

Kresge Library Home Page: http://www.bus.umich.edu/kresgelibrary/
<http://www.bus.umich.edu/kresgelibrary/>
My home page: http://www-personal.umich.edu/~cseeman/index.html
<http://www-personal.umich.edu/~cseeman/index.html>




________________________________

From: innopac-bounces at innopacusers dot org on behalf of Hewitt, Joseph W
Sent: Fri 1/27/2006 10:23 AM
To: innopac at innopacusers dot org
Subject: [IUG] questions about creating statistics reports in millennium



As a newbie at supporting Innovative products, I have learned that
acquiring circulation statistics from the same database will differ in
III. If you use Innopac, you get large numbers. If you use Millennium,
you get smaller, different numbers. We wish to use Millennium as much as
possible. My question is: how can I construct a circulation statistics
report that will deliver the same numbers I am currently getting through
Innopac for my statistical group?



More experienced librarians than I may know or be able to find out how
the query is stated by Innopac to retrieve information on circulation
activity by location for a particular timeframe. Millennium shows you
how the circulation statistics query is constructed through the GUI
wizard as you create it, but you are limited. If one could edit the
query in the box directly, a query could be constructed to replicate the
Innopac reporting process. Alas, in a turnkey system such as Innovative,
this capability does not seem to be available.



Our library does not have root access to the server.  We cannot look at
the database log for examples of how Innopac queries the system.



I was told that the Innopac numbers "don't match, won't match, and will
never match" by an experienced support person who also told me that III
recommends choosing one platform and sticking with it. If this is the
III policy, it seems patently ridiculous, especially if both systems
draw information from the same database.



My current results show the circulation numbers for last year in the
tens of thousands when extracted by Innopac, but only in the thousands
when extracted with the  Millennium client. Don't even think about Web
Management Reports.



Am I the only one wondering why the system will not or cannot provide
reliable, basic statistical information through its flagship
application-Millennium client?



Granted, I'm new and inexperienced and uninformed. So, if someone could
point me in a productive direction for developing reliable Millenium
circ statistics, I would be in your debt.



Thanks,

Joe



Joe Hewitt

Library Systems Specialist

Missouri State University





--- StripMime Report -- processed MIME parts ---
multipart/alternative
  text/plain (text body -- kept)
  text/html
---
--
This message was distributed through the Innovative Users Group INNOPAC
list
Public replies:  INNOPAC at innopacusers dot org
Update your subscription options:
http://innopacusers.org/mailman/listinfo/innopac




--- StripMime Report -- processed MIME parts ---
multipart/alternative
  text/plain (text body -- kept)
  text/html
---


--
This message was distributed through the Innovative Users Group INNOPAC list
Public replies:  INNOPAC at innopacusers dot org
Update your subscription options: http://innopacusers.org/mailman/listinfo/innopac




--- StripMime Report -- processed MIME parts ---
multipart/mixed
  text/plain (text body -- kept)
  application/ms-tnef
---