1 of 19

SCELC: Statewide California Electronic Library Consortium

Diversity in Shared Print Participant title holdings & retentions:

Minority Serving Institutions, � EAST & SCELC

PAN Annual Project Update, June 27, 2025

Jason Price, PhD

Director of License and Open Access Agreement Services, SCELC

EAST: Eastern Academic Scholars' Trust

This project was made possible in part by the Institute of Museum and Library Services (LG-252402-OLS-22). https://imls.gov/

Este proyecto ha sido posible en parte por el Instituto de Servicios de Museos y Bibliotecas, (LG-252402-OLS-22). https://imls.gov/

2 of 19

FAST Approach Overview

Planning Grant Research Questions (& Answers!)

Possible Collective Actions

Retention Commitment Patterns

Agenda

Susan

Here is the agenda for today:

- Following this welcome I will turn it over to Teri to provide an update on the status of the planning grant. We’ll focus on the 3 major goals and the projects that have been initiated in order to meet that goal, giving you an update on where we are.

As we review our progress against each of the goals, we will be identifying the challenges we have encountered and, based on progress to date, expectations we have for specific recommendations we will be making as part of our final report. Following EACH update on our progress against the goals, we will be asking for your feedback during a discussion period.

- As our final planning grant goal is focused on dissemination of the project’s results to the larger shared print and MSI library communities, we’ll take some time to seek your counsel on the most effective venues and formats for such dissemination

- And although we only have an hour for today’s call we’ll do our best to leave time for any final questions or comments at the end.

- With that, I’ll turn it over to Teri to kickoff the project update. Teri…

3 of 19

http://bit.ly/4lrNqH4

4 of 19

FAST

(Faceted Application of Subject Terminology)

Faceted metadata based on LCSH

More flexible than subject heading analysis

How is FAST metadata applied?

FAST Approach

Example LCSH Headings under FAST African Americans:

American literature -- African American authors
African American Churches
African American women doctoral students
Expatriation -- African Americans
Older African Americans

The collection analysis tool used in this work is OCLC’s Choreo Insights, which allows for the use of FAST subject headings. FAST headings are derived from Library of Congress

Subject Headings, but distill the over 26 million LCSH headings down to under 2 million FAST headings. This creates a kind of “shortcut” for subject analysis.

FAST headings are automatically applied by OCLC on any records which have LCSH headings.

(This means that 75% of materials in WorldCat are ‘fastened’). You can see an example Marc 650 field with a ‘fastened’ heading of “African Americans”. Using the FAST heading “African Americans” pulls in all related LCSH headings, including when the term is in a subdivision of the LCSH heading, for example 'expatriation – african americans’ or “American literature – African American authors” or even terms such as “Older African Americans”.

While FAST headings do suffer the same biases as LCSH headings, using FAST is more efficient, and just plain FASTer, than working with LCSH headings.

(80 seconds)

5 of 19

FAST is a vast improvement over LC Class/LC Range-based Diversity Assessment

EAST(121 Institutions))

DT1-3415: History of Africa

E184.5-185.98: Afro-Americans

PL8000-8844: African Languages & Literature

Range Recommendation From:

University of West Florida | University Libraries

Diversity Collection Audit & Assessment

https://libguides.uwf.edu/divassess/aa

Screenshots from OCLCs Choreo Insights

We also found that using FAST headings was a vast improvement over using LC class ranges for DEI work. Although the LC Class/Range approach identified 2x more books, the books matching FAST subject headings were across the LC spectrum and more relevant. Obviously if you are using Class Ranges to identify diverse materials, you are only going to find materials in those LC ranges, and miss out on titles related to your target group that fall outside of those ranges. As an example here, targeting african american materials, using FAST headings of african american/s, haitian american, and somali americans you find materials across the LC class spectrum, and in general materials those materials surfaced are more closely related to your target than you would get by using the narrower LC class search.

to just World History, Americican History Languages and Literature

6 of 19

Fast Headings used to Identify Books about Historically Marginalized Communities

7 of 19

Community Council Perspectives

Identifying books about communities of color

Should collection analysis use reductive techniques like FAST? Why or Why not?

Efficiency and Scalability

Shortcomings

lack of human oversight, SMEs (though that may be limiting)
who decides on matching terms?
does not identify authors from Communities of Color
lack of granularity in the approach

8 of 19

Research Questions

Do MSI book collections reflect the communities they serve?

Do MSIs own titles about their communities that are not held by other shared print participants?

Do MSI retention commitments reflect the community they serve? �

What proportion of titles about these communities have been retained?

(4)

All three of these questions were recommended by our IMLS planning grant reviewers.

�Notice that they all rely on identifying books that are about communities of people.

�We refer to these communities as “historically marginalized”,

fully recognizing that this phrase is a woefully inadequate way to refer to these communities, since the effects of systemic and interpersonal racism they experience are not just marginalizing and not just historic.

�In a similar fashion, we acknowledge that our approach to identifying books about these communities is far from ideal, both because we are using a narrow definition that is dependent on the underlying LCSH headings, and because we are outsiders, who are far less qualified to determine aboutness than are members of the communities themselves.

�Our hope is that these significant shortcomings are tolerable in that working with them for now enables a small step forward in improving diversity assessment of large collections.

�(We benefitted greatly from the guidance of our MSI community council, for improvements to our terminology and report)

Jason

Questions that our grant reviewers recommended we get answers to.

We designed the study around them and are working hard to get the best possible answers we can.

At the same time, we feel like any accurate answers no matter how limited are a significant step forward.

And rather than start with the limitations, we decided to present the preliminary results right up front, so here we go...

9 of 19

(1) Do MSI book collections reflect the communities they serve? YES

10 of 19

(2) Do MSIs own books about their community that are not held by other SPP participants?� YES

11 of 19

(3) Do MSI retention commitments reflect the communities they serve? � Only to the extent that their collections reflect the community

12 of 19

(4) What proportion of titles about these communities have been retained?

42-59% by MSI Community

13 of 19

Is this retention rate surprisingly low?

Is it a failure of our retention models or strategy?
Can it be explained by differences between the two sets of titles?

Dates? Special collections? Narrow vs Widely held?

Does it differ from the overall retention rate?
What other possible explanations are there?

15 of 19

Community Council Perspectives

Preserving collections worldwide

Which possible program initiative(s) would provide the greatest benefit for your institution or community and why?

Ensure 100% retention of these titles (including those out of date scope)
Retain More Copies

in many cases, they were published in low numbers to begin with
“Play a role in safeguarding unique materials about underrepresented communities for future generations despite the political challenges they face today”

16 of 19

Collective Actions?

What can our shared print communities do to make more of books about communities of color available to MSIs specifically and to researchers everywhere?

Some Possibilities

Ensure 100% retention of these titles (!!!)
Cultivate & share lists for voluntary commitments
Consider retaining more (or even all?!?) copies
Focus on lending and access
Recruit libraries that own unretained titles about communities of color