[Eril-l] Detecting missing titles in vendor collection metadata

Wed Jan 2 08:23:25 PST 2019

Hello, Kathleen,
You asked " Do you mind my asking how you were able to detect the missing
records?" [from the OCLC collection missing Sept 2018 titles from Oxford
Scholarship Online]

When I have free time (during semester breaks and summer), I will compare
Knowledge Base data with data from vendors.  I use vendor spreadsheets from
many sources.  The sales title-price lists are often the most current and
complete.   Also I have checked vendor supplied back-end data, such as
 vendor supplied MARC records, or vendor "support" supplied title lists.
I've even screen scraped patron interface searches, and found titles not
listed in the sales or back-end title lists.  I've used older entitlement
lists kept by my library, checking for "perpetual" access, especially when
my library has cancelled titles yet should have access to back issues.

I usually select more expensive collections.  And I try to fix, or work
with vendors to fix the problem, in a way that all libraries will receive
the corrected data, when possible.

Regarding software, I use Excel to compare different lists, and MARCedit to
extract spreadsheets from bib records.  I've talked with others who use MS
Access or OpenRefine.

Overtime, as an ERIL community, we need to develop more systematic ways of
monitoring and improving vendor and KB data.  In the five years I've been
examining collection data, I've made progress on collections important for
the libraries who employee me, yet there's a steep, trial-and-error,
learning curve, and little written.

(A significant problem which I've not had resources to tackle are journal
previous titles when the vendor only supplies the current title.  One large
vendor, Springer, only supplies current titles, and some other vendors
too.  For example, the title Soviet applied mechanics =  Prikladnaya
mekhanika (1966-1992) was picked up by Springer, and the title changed to
International Applied Mechanics.  But Springer does not include "Soviet
applied mechanics" in their metadata, so a patron with a footnote citation
for an older article thinks our library does not have access; and even ILL
for that article is difficult.  There are hundreds of Springer previous
titles which are not discoverable.)

Metadata for collections needs greater staffing, much discussion and
cooperative work, and perhaps improved standards.   There's so much more to
discuss, but I'll close with this quote from:

Reengineering the Library: Issues in Electronic Resource Management, edited
by George Stachokas. Chicago: ALA editions, 2018.

"The development of community-managed knowledge bases is a very promising
approach for the maintenance of an expanding volume of collection sites,
eBooks, and other resources that are now taking a major portion of our
acquisitions budget. …
Libraries can be more successful in electronic resources management if they
work toward the greater good. No single library or even a statewide
consortium can meet the many demands of providing access to electronic
resources. Libraries must be part of a larger initiative to combine our
efforts a the national level, and preferably even at the international
level." Richard Guajardo. p 56.

------------------------------
Date: Fri, 21 Dec 2018 10:41:15 -0500
From: Kathleen Folger <kfolger at umich.edu>
To: J Siemon <jsiemon2002 at gmail.com>
Cc: "Eril-l at lists.eril-l.org" <eril-l at lists.eril-l.org>
Subject: Re: [Eril-l] OCLC collection missing Sept 2018 titles from
        Oxford Scholarship Online
Message-ID:
        <CAK6MXLXJKxz9FgOGrwe9Hd9wbJqo5WbR5VQmTqvrSZbhFbn1gg at mail.gmail.com>
Content-Type: text/plain; charset="utf-8"

Hi Jeff,

Thanks very much for the message and the links to your presentations. I've
been concerned about this issue as well, so I'm looking forward to delving
into them.

Do you mind my asking how you were able to detect the missing records? Do
you have a process in place for comparing the number of records received to
number of titles purchased? For the most part, we simply accept the records
available and don't do a check to try to confirm we have records for every
title included in a collection we've purchased. Often, especially for
frontlist ebook packages, that information isn't even available from the
publisher so we end up relying on our subject specialists or users to
identify missing records. That's not a great strategy, so I'd be interested
in hearing about ways others are trying to address the issue. Thanks!

-Kathleen
_________________________________________
Kathleen M. Folger, Electronic Resources Officer
University of Michigan Library
312 Hatcher North
Ann Arbor, MI 48109-1190
V:(734) 764-9375
F:(734) 764-0259
kfolger at umich.edu
<kfolger at umich.edu>

My pronouns are she, her, hers ? what are yours?

On Fri, Dec 21, 2018 at 9:58 AM J Siemon <jsiemon2002 at gmail.com> wrote:

> OCLC Knowledge Base collection Oxford Scholarship Online Religion =
> OUP.osoReligion is missing the Sept 2018 Religion titles from Oxford
> Scholarship Online, and three older titles.  OCLC and Oxford are both
> excellent organizations, yet sometimes even reliable vendors make
mistakes.
>
> This is a good example of a larger issue I've been discussing at
> conferences for the last couple years.  Libraries need to devote
> significantly more staff resources to reviewing collection metadata.
> Libraries hire thousands of catalogers, who spend time reviewing MARC data
> for *individual* titles.  I'm convinced that some of these people/staff
> resources need to be reallocated to review the accuracy of *collection*
> level data.  Book and journal budgets have shifted toward purchasing
> collections of eResources (instead of individual titles).  Library
staffing
> needs to shift in a similar manner, toward reviewing and correcting
> collection level metadata in addition to individual title data.  When
> collection level data is wrong, tens, and sometime hundreds of e-books or
> e-articles are not available to our patrons.
>
> I've contacted support staff at both organizations.
> I don't know whether Oxford neglected to send this data or OCLC neglected
> to load these titles.
>
> In any case more library staff need to be devoted to noticing these
> collection level metadata issues, and trouble-shooting with vendors to
make
> the corrections.  See my article from the Charleston Conference, 2017,
>  "You May Own It . . . But Can They Find It? A Panel Discussion: Part 3 of
> Panel Presentation: Collection-Level Cooperative Cataloging"
> https://dx.doi.org/10.5703/1288284316663 or some of my presentations at
> slideshare https://www.slideshare.net/jsiemon/presentations
>
>
>
> All the best,
>
> Jeff Siemon
> _______________________________________________
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.eril-l.org/pipermail/eril-l-eril-l.org/attachments/20190102/779123f0/attachment.html>