OhioLINK Collection Analysis Project

1
OhioLINK Collection Analysis Project
OCLC Members Council
Preliminary Analysis
21 October 2008
Ed ONeill, OCLC Research With support and
contributions from Julia A. Gammon, University
of Akron Anne T. Gilliland, Ohio State University
(Formerly OhioLINK)
2
1987 Library Study Committee Report

• Key Recommendations
• Create a book depository system
• Create a statewide electronic catalog
• Appoint a steering committee

3
OhioLINK Planning Paper

• Coordination in purchasing of shared collections
• Expanded access to electronic information
• Improved access to information infrastructure
• Promotion of scholarly communications
• Improved economies in purchase of electronic
  resources

4
Who is OhioLINK?

• The State Library of Ohio
• 5 ARL Institutions
• 11 Universities
• 44 Colleges
• 15 Community Colleges
• 28 Branch campuses
• 5 Depositories
• 3 Museums and other independent cultural
  institutions
• 20 Off-campus hospitals and medical centers

5
What is OhioLINK?

• Shared catalog with patron initiated borrowing
• 600,000 Users
• 47.6 million books and other library materials
• Millions of electronic articles
• 12,000 electronic journals
• 140 electronic research databases
• 40,000 e-books
• Thousands of images, videos and sounds
• 17,500 theses and dissertations from Ohio
  students

6
Research Project

• Joint study by OhioLINK, OhioLINK members,
  OhioLINK Collection Building Task Force (CBTF)
  and OCLC Research
• Much of the planned analysis is new and
  untested not all of the analysis will be
  successful
• This project is distinct from OCLCs collection
  analysis service

7
Distinctive Aspects

• Size and scope of collections
• Use of local holdings information
• Number and variety of institutions
• FRBR (Functional Requirements for Bibliographic
  Records)
• Application of Audience Level

8
Project Goals
For the book collection

• To reduce unnecessary duplication
• To increase local collection development
  activities
• To expand the amount spent on cooperative
  acquisitions
• To strengthen the collective collection

9
Books What we hope to learn

• Is the OhioLINK collection getting more diverse?
• Is duplication increasing or decreasing? How
  much is justified?
• What does the OhioLINK collective collection
  look like?
• What books didnt we acquire?
• Does the 80/20 rule apply?
• Are the acquisitions budgets effectively
  allocated?
• What is the average age of the books by subject?

10
What we hope to learn (Cont.)

• Does the size of core collections vary by
  subject?
• What is the half-life of books in a particular
  subject areas?
• Does circulation correlate with the
  strengths/specialties/programs?
• Are the sciences really not using books?
• Does circulation correlate with number of
  copies? With WorldCat holdings?
• Do usage pattern vary by institution?
• Are the ARLs different or just larger?
• What books should be in the depositories?

11
OhioLINK Circulation Data

• Item No.
• OCLC No.
• Title
• LCCN
• Location Code
• Status Code
• Circulation
• Renewals
• Accession date
• Date of Last Use
• ISBN
• Source

i25878591 45207959 The infinite / A.W.
Moore 00051722 bc - 5 1 8/3/2001 8/23/2004 0415252
857 (pbk.) Akron
12
Accumulative Circulation Data

• Makes comparison difficult An item with high
  circulation may be currently be little used, i.e.
  Word97
• To obtain current circulation rates, Pre/post
  images will used
• The first data set of circulation data was
  collected in the Spring of 2007
• The second data was collected this Spring (2008)
• From the second set of data, the circulation for
  the past year can be determined

13
Data Collection Schedule

• First Snapshot April - May, 2007
• Second Snapshot April May 2008
• Validation of circulation policies July
  October 2008

14
WorldCat Linking
The OCLC Number is used to link the circulation
records to the corresponding bibliographic record
in WorldCat

• For records with an obsolete OCLC No. the
  obsolete OCLC No. is replaced with current OCLC
  No.
• For records without an OCLC No. which had either
  a unique LCCN or ISBN that number is used to
  identify the corresponding OCLC No.
• Records lacking any standard number could not be
  validated and were excluded from the study

15
Validation

• Validating link
• The title from the OhioLINK circulation record
  was compared to the title from the WorldCat
  record
• If the title from the circ record was similar to
  the title in the WorldCat record, the record was
  validated
• Records with dissimilar titles were not be
  validated and were excluded from the study
• Determining material type
• Only books and manuscripts were included
• Material type was based on fixed fields codes in
  the WorldCat records (bib lvl m and type a or
  t)

16
Validated OhioLINK Circulation Data

• Records Received 33,146,008
• Records Validated 30,718,454 (92.7)
• Validated Books 27,002,190 (81.5)

17
FRBR Group One Entities
18
Humphry Clinker Example

• 53 OhioLINK libraries hold the work
• 1 English language expression
• 48 Different manifestations

19
Lots of Different Manifestations
20
Most Common in OhioLINK
OCLC No. 358955 52 Copies 27 Libraries
Not held by the University of Cincinnati
University of Cincinnati does hold 9 other
manifestations
21
Audience Level

• Audience level identifies the audience for which
  the book or other library resource is suitable
• Audience level is inferred from the type of
  libraries (ARL, Academic, Public, School) that
  have acquired the resource using the library
  holdings data from WorldCat
• The audience level ranges from 0.0 (Juvenile) to
  1.0 (Scholarly)

22
Audience Level Examples
A collection can be characterized by average
audience level of its resources.
0
1
23
Library Organizational Structure
The large universities are complex organizations

• Multiple administrative units
• Many different physical locations
• Branch campuses
• Depositories
• Independent cultural institutions
• Off-campus hospitals and medical centers

24
Library Organizational Structure

• Campuses, independent cultural institutions, and
  depositories are treated as top (first) level
  units
• Independent administrative units (if present)
  within the campus
• Separate libraries (if present) within an
  administrative unit
• Distinct collections with unique location codes

25
Multi Level Structure
Top level units Individual campus,
depositories, and external organizations
(Museums, Centers, Hospitals)
Second level units Separate administrative
units university libraries, law, medicine,
etc.) or distinct library units.

• Third level units Distinct library units.

Location codes The codes used within OhioLINK
to identify to location of the individual
items. Over 4,200 different location codes
were found one institution alone used 556
different codes)
bccco, bccct, bccir, bccm, bcgd, bcgdo, bcmu,
26
Holdings Detail
27
Three Level Structure for Akron
28
Location Codes Mapping
29
Caution!

• The project is still in progress and the data
  analysis is incomplete
• Results are preliminary revisions and
  corrections will occur

30
General Information
31
Languages
Additional columns include statistics for German,
French, Spanish, Russian, Chinese, Italian,
Japanese, Latin, Hebrew, Polish, Greek, and Arabic
32
Subjects
The subject analysis included 24 primary
subjects a more detailed subject analysis with
approximately 500 subject areas will included in
the final analysis
33
24 Primary Subjects
Agriculture Anthropology Art and
Architecture Biological Sciences Business and
Economics Chemistry Computer Science Education Eng
ineering and Technology Geography and Earth
Sciences History and Auxiliary Sciences Language,
Linguistics, and Literature
Law Library Science, Generalities, and
Reference Mathematics Medicine Music Performing
Arts Philosophy and Religion Physical Education
and Recreation Physical Sciences Political
Science Psychology Sociology
34
Age
Statistics on 20 different age groups are provided
35
Collective Collection What Do We Have?
How many items do we have? What languages do we
have? How old are they? How many are unique? In
what subjects? How many copies do we need?
36
Most Held

• Libraries 68
• Copies 109
• Circulations 99

37
Most Copies

• Libraries 12
• Copies 9,542
• Circulations 9

38
Most Circulated

• Libraries 6
• Copies 92
• Circulations 6,023

39
Holdings vs. Active Collection
40
Subject Distribution
41
Circulation by Subject
42
Hot Subjects

• Computer Science (QA 75-76)
• Women, Feminism, Life Skills, Life Style (HQ
  1101-2044)
• Medicine Special Subjects (R 690-920)
• Buddhism (BQ)
• Nursing (RT)
• Broadcasting (PN 1990-1992)

43
Language Distribution
24,386,814
44
Circulation of Non-English Materials
Average per Item Circulation
45
Circ. Rate by Institution Type
2.3
3.6
2.3
Circulation
1.7
ARL
Univ.
Colleges
CC/ Branches
46
Usage Distribution
of Circulation
of Books
47
Annual Collection Growth
Max 114,375 (2000)
No. of Manifestations Added
Publication Date
48
Duplication Rate
Average No. of Copies
Publication Date
49
Duplication by Subject
50
Conclusions?

• Only first phase of data analysis complete
• Additional and more reliable statistics will be
  available after the next phase
• Preliminary results
• Duplication rates are steady
• The 80/20 rule may be closer to 80/10
• Limited use of non-English materials
• Books are still being used in the Sciences
• Circulations rates vary greatly by subject,
  institution
• To be continued .

51
Since the beginning Collaboration continues

• Continue the cooperative spirit in collection
  building through
• Increased the understanding of the distribution
  and use of library resources in the collective
  collection
• Utilization of distribution and usage information
  as a guide for building collections
• Open and wide dissemination of statistics and
  analysis

52
Questions?
Ed ONeill OCLC Research oneill_at_oclc.org 614-764-
6074 This presentation is available on the OCLC
Web Site at http//www.oclc.org/memberscouncil/me
etings/2009/october/researchsg-oneill.ppt