Data to Decisions: Shared Print Retention in Maine

Data to Decisions:
Shared Print Retention in Maine
Deb Rollins, University of Maine
Becky Albitz, Bates College
Charleston Conference November 8, 2013

www.maineinfonet.org/mscs/

What is Shared Print Anyway?
 Holding libraries commit to retain designated materials for a

specified time period so that partner libraries can rely on
their continued availability.

2

www.maineinfonet.org/mscs

U.S. Print Retention Projects
 West Storage Trust – Western US – Journals
 CIC – Large Midwest universities – Journals
 ReCAP – Columbia, Princeton, NYPL
 HathiTrust
 Scholars Trust -AESERL & WRLC – Journals
 Northeast Regional Library Print Management Project –

Monographs & Journals
 Maine Shared Collection Strategy – Monographs & Journals
 Center for Research Libraries – PAN network
3


Drivers for Print Retention
 Space issues – lack of it!
 Budget cuts – where has all the money gone?
 Cost per usage – why did we buy this again?
 Availability of electronic resources – paperless library?

4


Building on Collaboration and Trust
 MaineCat has encouraged resource sharing
 State-wide delivery – 1.25 millions items/year
 Colby, Bates, and Bowdoin (CBB) are consciously building a

shared collection of new print materials and e-resources
 Trust in commitments and continued access

5


Project Background
 Originated with the Larger Libraries Group
 Most libraries were running out of space and unlikely to get

additional storage
 Wanted a shared approach to managing legacy print
collections for the long-term
 Looking to be leaders in the print collection space

6


Project Partners

7


Grant Proposal
IMLS grant of $821,065 to create a shared print collections
strategy:
 Create a collection analysis system
 Include large scale digital collections (HathiTrust and
Internet Archive) when determining what to keep in print
 Develop a strategy to make retention decisions at scale
 Integrate E-book-On-Demand and Print-On-Demand

8


Grant Summary
 Monographs and journals (Government documents are






9

excluded)
Which titles should be retained long-term and by whom
Libraries may choose to discard, once retention decisions are
made, although downsizing is not the predominant focus
Provide a framework for other libraries to join once the
initial grant period is complete
Disclose our retention decisions locally and to the world
Be part of the national conversation


Project Management
 Project Team: Program Manager, Technology Director, Project PIs

& Systems Librarian

 Directors’ Council
 Collection Development Committee
 Technical Services Committee
 National Advisory Board

10


Governance and Business Model
 A Memorandum Of Understanding has been developed to

guide the ongoing work
 15 year retention commitment
 MOU and commitments reviewed every 5 years
 Executive Committee will provide governance
 Collections and Operations Committee will determine

retention, holding disclosure, and access/delivery
 Different levels of membership
 Collection Holders
 Collection Builders
 Supporting Members

11


MSCS Goals
 Identify long-term retention commitments from libraries
 Implement on-demand services in union catalog
 Define sustainable business model for beyond grant &

current partners

12


MSCS Guiding Questions
 What monographs should be designated for long-term

retention?
 What is an equitable and/or common-sense distribution of
retention responsibilities?
 What monographs are candidates for incorporating into
POD/EOD services by virtue of HathiTrust or Internet
Archive public domain availability?
 What monograph copies could optionally
be deselected, once retention decisions have been
finalized?
13


Data – Data Starting Points
 How many copies of a particular work are owned by partner








14

libraries?
How many of those are circulating copies?
How often has the title circulated? What was the last circulation
date?
How many titles/copies are uniquely held in the group? In Maine?
In WorldCat?
How do subject strengths compare across the group?
Which titles are represented in HathiTrust, Internet Archive?
Overlap between general and special collections
Others to be determined from combined data set

The Data - Where is it?
 Local
 5 ILS catalogs
 8 libraries
 Innovative Interfaces, Inc. (III)

 State
 MaineCat
 INN-Reach catalog of more than 100 Maine libraries

 National
 OCLC
 HathiTrust
 Internet Archive

15


Data – The Details
 OCLC reclamation
 Cleaned up holdings and OCLC numbers
 Facilitated match across partners

 Local outputs (to SCS - Sustainable Collection Services)
 Monographs
 Exclusions

 Local outputs (to Systems Librarian)
 Serials

16


The Data – Extracted Fields
 Complete bib data
 OCLC number

 Item record data e.g.
 Call number
 Location
 Usage counts
 Last checkin
 Circ status

17


Data – SCS Actions
 Additional data cleaning—normalizing, de-duping, and filling

in missing data
 Matched titles to external data sources—OCLC WorldCat
(U.S. and State Holdings), HathiTrust Public Domain and InCopyright items, and Internet Archive
 Consulting support
 Data reports

18


High Level View of the [Monograph] Data
Bib records – Unfiltered

Item records – Unfiltered

Libraries

2,958,905

3,420,061

9

Bib records – Filtered
2,920,014
(circulating titles
2,719,754)

Item records – Filtered
3,374,574

Unique Titles – Filtered
1,754,598

19


By “titles" we can mean two different
things
1. Title Set

Bates

Bowdoin

Colby

Maine SL

Portland PL UM-Orono

2. Title Holding
20


USM

Circulation Counts –
Circulating Titles Only
MCSC Title‐Holding Counts
1

All Libraries

All Filtered Title Holdings - CIRCULATING
TITLES

%

2,719,754 100%

Circulation Counts
2

845,939

31%

3

Total Charges = 1

466,371

17%

4

Total Charges = 2

303,588

11%

5

Total Charges = 3

206,610

8%

6

Total Charges = 4 to 9

511,040

19%

7

Total Charges = 10+

386,206

14%

8

Last charge after 2010

357,660

13%

9


671,815

25%

10

21

Total Charges = 0


841,009

31%


WorldCat Counts - US
14
15

145,296

5%

0‐9 Holdings in USA ‐ FRBR

95,571

3%

16

10‐19 Holdings in USA

94,162

3%

17


59,386

2%

18


213,827

7%

19


146,868

5%

20


290,443

10%

21


222,700

8%

22

100‐199 Holdings In USA

507,552

17%

23

100‐199 Holdings In USA ‐ FRBR

422,454

14%

24

200+ Holdings in USA

1,668,732

57%

25
22


200+ Holdings in USA ‐ FRBR

1,973,033

68%


Overlap within MSCS Group
40

Unique Holding in group

1,118,151

38%

41

Title‐holdings in 2 libraries

684,395

23%

42


462,446

16%

43


325,959

11%

44


190,215

7%

45


82,224

3%

46


40,179

1%

47


15,550

1%

23


*http://sustainablecollections.com/weed-feed/2013/9/25/the-scs-monographs-index.html


99 titles are held by all 9 MSCS Institutions
Pub
Year

MSCS
Total Circs

Nickel and dimed : on (not) getting by in America / Barbara Ehrenreich.

2001

1,906

Gilead / Marilynne Robinson.

2004

1,582

The diary of a young girl : the definitive edition / Anne Frank ; edited by Otto H.
Frank and Mirjam Pressler ; translated by Susan Massotty.

1995

1,064

Seed of Sarah : memoirs of a survivor / Judith Magyar Isaacson.

1990

859

The lobster gangs of Maine / James M. Acheson.

1988

713

In the hands of Providence : Joshua L. Chamberlain and the American Civil War
/ by Alice Rains Trulock.

1992

657

Bowling alone : the collapse and revival of American community / Robert D.
Putnam.

2000

655

A distant mirror : the calamitous 14th century / Barbara W. Tuchman.

1978

570

Liberty men and great proprietors : the revolutionary settlement on the Maine
frontier, 1760‐1820 / Alan Taylor.

1990

552

1988

515

Title/Author

25
Collected works / Flannery O'Connor.

1,200,000

MSCS Circulating Title-Holdings by
Holding Level – Circulation Levels

1,000,000

800,000

374,062

600,000

Zero Circulations
1-3 Circulations
4 plus Circulations

204,219

267,658

403,284

341,231
400,000
232,054
200,000

295,425

393,391
208,430

1

2

3+

Number of MSCS Libraries Holding Title

Data – Split into Two Steps
 Step 1 “Not widely held” title-sets
 Title held in 1 or 2 MSCS libraries
 Publication year < 2003

 Step 2 “Widely held”
 Titles held in 3 or more MSCS libraries
 Publication year < 2003

27


MSCS Title-Holding by Holdings Level
1,800,000
1,600,000
1,400,000

Not Widely Held Titles
Step 1

1,200,000
1,000,000

Widely Held Titles
Step 2

800,000
600,000

1,655,421
1,064,333

400,000
200,000
-

1-2

3+


Step 1 – Further Divisions
 Limited to publication date <2003
 Commit To Retain – CTR – if :
 Any circulation, internal, or reserve use OR
 “local interest” title-sets OR
 Special Collections items OR
 Specific edition held in 9 or fewer libraries in the U.S.

 Needs Further Examination – NFE – if:
 Zero circulations

29


“Local Interest” Rules Applied
Rule 1: General: ‘Maine’ will be searched in title, author, series, and all subject fields
Rule 2: Published in Maine: Search bib records for indication of Maine as place of
publication
Rule 3: Maine Author or Artists: Location code meaut in a Maine State Library
record, Authors, American—Maine in a subject field, Artists, American—Maine in a subject
field
Rule 4: Maine Local History: Title is classed in: F 16-30 (Maine History), 917.41 &
974.1 & meanx
Rule 5: Major Colleges & Universities: Keywords/phrases searched e.g. Bowdoin
College
Rule 6: Industries: Keywords/phrases searched (combined with Maine) e.g. Paper
Industry
Rule 7: Marine & coastal studies: Keywords/phrases & classifications e.g. QH 92-92.2
Marine Biology (Atlantic Coast)
Rule 8: Native Americans: Keywords/phrases e.g. Abenaki
Rule 9: Places/Populations: Keywords/phrases e.g. Acadia
Rule 10: Religious groups: Keywords/phrases e.g. Free Will Baptists
30


Scenario One – Needs Further Examination / Commit to Retain
1,800,000


1,600,000

Published After 2003
(removed from Step 1) 186K

1,400,000

Step 1
Needs Further Examination
392,382

1,200,000
1,000,000

Widely Held Titles

800,000
600,000

Step 1
1,655,421
Commit to Retain
1,076,188

Step 2
1,064,333

400,000
200,000
‐

1‐2

3+



Total Commitments & Needs Further
Examination for Step 1
LIBRARY

COMMITMENT TO RETAIN
Titles

Bangor Public Library

Items

NEEDS FURTHER EXAMINATION
Titles

Items

147,490

177,195

40,582

43,182

9,688

18,095

9,536

11,921

Bates

129,168

142,603

53,403

57,043

Bowdoin

161,498

202,550

95,497

108,888

Colby

124,178

142,617

49,005

53,516

Maine State Library

43,532

53,726

6,458

7,782

Portland Public Library

78,065

97,133

3,678

4,316

University of Maine Orono

276,784

307,202

119,793

130,218

University of Southern Maine

105,785

117,074

14,430

15,392

Bangor Theological Seminary

32

ALL ELIGIBLE TITLES

1,258,195
392,382

1,076,188

432,258

“Needs further examination” group
 1-2 MSCS title holdings
 0 recorded uses
 37% in copyright in Hathi
 6% public domain in Hathi
 6 % digitized in Internet Archive (no Hathi overlap)
 51% not digitized
 Made decision NOT to Commit to Retain these

33


Drowning in the Data
 Too many CTR?
 Too many lists
 Piles of objectionable CTR books event
 Publishers (29,331 title holdings had commitments reversed)

34


Step 2
1,800,000
1,600,000
1,400,000
1,200,000

No CTR
421,713

Widely Held Titles

1,000,000

Published after 2003
800,000
600,000

Step 1
1,655,421
Commit to Retain
1,046,857

Needs Further Examination
1,064,333
In-Scope for Step 2

400,000

- Titles published < 2003
-Circulating Titles
-10+ US holdings

200,000
‐

1‐2

3+



Title-Holding Counts by Title-Set Usage Levels (uses per title-holding)
250,000

IN-SCOPE TITLE SETS
Publication year < 2003
Three plus libraries holding
Circulating titles only
No special collections
No Hathi public domain
10+ US holdings

216,401 (28%)
200,000

150,000

136,711 (18%)

Title-Holdings

132,924 (17%)

Title-Sets

100,000
81,336 (10%)

70,691 (9%)

58,316
50,000

38,353

54,911 (7%)

53,871 (7%)

34,653

32,349 (4%)

20,728

9,836

17,679

13,504

13,662

7-10

10-15

> 15

Zero

0-1

1-3

3-5

5-7

Uses per Title-Holding per Title-Set (SCS calculation)


Step 2 Questions
 Do we need to retain a minimum or maximum number of

title-holdings per title-set?
 How to allocate holdings responsibility?
 Circulation policies
 Library type
 Subject strengths
 CTRs made in Step 1

 We are still discussing how these elements work together to

create an equitable model.

38


Data – Serials are Different
 All records coded format “s”
 Analysis by MSCS Systems Librarian
 Includes many book series as well as journals/magazines
 Inconsistent across MSCS institutions

 Total number of serials records 37,887
 Total number of deduped serials records 27,912

39


MSCS Goals for Serials
 Goal is not to duplicate existing efforts
 Relative safety of publisher digital collections
 Digital services e.g. Portico

 Act as ‘good steward’ locally and nationally
 CTR “local interest” titles
 CTR “not widely held” titles
 CTR Special Collections titles

40


Serials Decisions
 27,912 title-sets (book series and journals)
 Not CTR ~ 64%
 4,198 digital collections overlap
 14,771 >50 in OCLC
 CTR <36%
 8,943 or less
 6,134 local interest and Special Collections match
 4,439 <50 OCLC
o MSCS overlap only 3% within this set (~130)
o To individual library review

41


Serials <50 OCLC review example

 Opportunity to review lists
 UMaine 1,977 titles--reduced to CTR 1,610 titles
 Bates 315 titles—reduced to CTR 254 titles
42


Disclosing Retention Decisions:
The Why
 The principle of the common good
 Participating as a partner in the national print retention

world
 Allows discards where appropriate
 Local workflow

43


Disclosing Retention Decisions:
The How
Disclosure has not been a straightforward process!
 National and International – OCLC WorldCat
 OCLC Shared Print symbol

 State – Central union catalog: MaineCat
 Commitment note pulled from OCLC

 Local –Five catalogs
 MARC 583 Action Note

44


Display in WorldCat.org


Display in OCLC FirstSearch

46


Display in Union Catalog


Display in Local OPAC

48


Display in local staff view

49


Ebook-On-Demand/Print-On-Demand
Service Model
 MSCS libraries are using large-scale digital collections like

the HathiTrust and Internet Archive in the management of
their print collections:
 Developing criteria for relying on digital copies as surrogates
 Integrating 1.6 million HathiTrust Public Domain title records

and Google Books links into our union catalog
 Print-On-Demand request service

50


MSCS Group Collection Summary:
HathiTrust and Internet Archive Overlap

51


Service Delivery Model

52


How Do You Replace the King?

53


What are our responsibilities moving
forward?
 If a CTR title is missing?
 Attempt to replace at a reasonable cost
 Request the transfer of the title from a non-participating library
 Ask Collections and Operations Committee to remove

retention commitment.
 If a title is damaged?
 Box and reshelve
 Attempt to replace
 Request a transfer
 Ask for retention commitment removal

54


Storage
 MSCS chose a distributed model rather than a centralized

storage facility
 Ownership and storage will remain with the owning library
 There are no different circulation or Interlibrary Loan rules or

workflows for titles with retention commitments

55


Lessons Learned
 Things won’t go as planned
 Perfection is not possible
 Need for a dedicated project manager position
 Public libraries are different!
 Libraries can cooperate

56


Legacy of MSCS
 Cooperative collection development amongst partners
 Extend membership to other Maine libraries
 Part of wider shared print community
 Extend model to other projects

57


Thank you!
drollins@maine.edu
ralbitz@bates.edu
www.maineinfonet.org/mscs/
@MESharedColls

58


Data to Decisions: Shared Print Retention in Maine

Recomendados

Recomendados

Más contenido relacionado

La actualidad más candente

La actualidad más candente (20)

Similar a Data to Decisions: Shared Print Retention in Maine

Similar a Data to Decisions: Shared Print Retention in Maine (20)

Más de Maine_SharedCollections

Más de Maine_SharedCollections (19)

Último

Último (20)

Data to Decisions: Shared Print Retention in Maine