Using Evidence in Practice

Fostering Evidence-Grounded Dialogue in a Multi-Institutional Digital Library

Leah Duncan

Digital Collections Librarian

Louisiana State University Libraries

Baton Rouge, Louisiana, United States of America

Email: lpowe17@lsu.edu

Elizabeth Joan Kelly

Library Web and Applications Development Administrator

LOUIS: The Louisiana Library Network

Baton Rouge, Louisiana, United States of America

Email: elizabeth.kelly@laregents.edu

Sophia Ziegler

Head of Digital Programs and Services

Louisiana State University Libraries

Baton Rouge, Louisiana, United States of America

Email: sziegler1@lsu.edu

Received: 1 Apr. 2022 Accepted: 14 Apr. 2022

2022 Duncan, Kelly, and Ziegler. This is an Open Access article distributed under the terms of the Creative Commons‐Attribution‐Noncommercial‐Share Alike License 4.0 International (http://creativecommons.org/licenses/by-nc-sa/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly attributed, not used for commercial purposes, and, if transformed, the resulting work is redistributed under the same or similar license to this one.

DOI: 10.18438/eblip30139

Setting

This report presents the findings of a survey of Louisiana-based librarians, archivists, and museum professionals’ processes for digitization selection and digital collection outreach and assessment. Survey participants were administrators at cultural heritage institutions that contribute to the Louisiana Digital Library (LDL), a state-wide resource for sharing digital heritage content.

The survey was part of a larger, grant-funded project, “...And 25 of our closest friends: The Louisiana Digital Library as Community-Focused Data (“The LDL,” n.d.) awarded by Collections as Data: Part to Whole (n.d.), with funds made possible by The Andrew W. Mellon Foundation. The main goal of the project team was to bring together LDL professionals as a community of practice and explore the policy, ethics, and implementation of reconceptualizing the LDL as data. The LDL community had never before gathered in such a way because, structurally, the LDL is a decentralized network of contributing institutions. The survey team is composed of librarians who work with the LDL community to provide ongoing training, support, and development, while individual institutions have autonomy over their own collections.

Problem

This decentralized model means there is not a unified vision for the role of the LDL in representing Louisiana cultural heritage. Different institutions pursue different goals related to the digital library’s research value, representational inclusivity, and scope. We also did not understand what steps, if any, individual institutions take to assess the use of their own content, and if that information ever influences what they choose to digitize and add to the library.

Furthermore, participating institutions vary widely in their size, staffing, and endowments, and we did not know how many professionals at each institution work with the LDL, or how staffing levels influence digitization and collection assessment. Without this basic information it is difficult to develop vision statements for the digital library, facilitate sustained planning sessions with our community of content administrators, or equitably distribute LDL resources.

Evidence

Using Qualtrics, we distributed a survey to gain insight into the current practices of LDL institutions in relation to (1) digitization selection, and (2) assessment of collection content, use, and reuse (Ziegler et al., 2020). The authors distributed the survey to 27 LDL content administrators who served as the primary decision-makers regarding their institution’s digital collections. In total, 22 content administrators responded, representing 15 academic libraries, three public libraries, two museums, one cultural heritage institution, and one respondent who chose not to disclose their institution type.

We found that digitization selection staffing and processes vary widely. Fifty-five percent of respondents had one to two full-time employees contributing to digitization initiatives, and 27% have more than two. Nine percent had less than one full-time employee contributing. The remainder of respondents did not have ongoing digitization initiatives. Regarding workflow, 36% of responding institutions had certain individuals who chose what gets digitized, 25% based their digitization selection decisions on community and patron requests, 7% had a digitization selection committee, and 7% based selection on grant funding and donations. The remaining 25% did not have a solidified process for deciding what would be digitized from their collections. Sixty-one percent of respondents expressed interest in modifying their selection strategy.

Most respondents indicated they performed some form of collection assessment. Fifty-eight percent assessed the content and/or scope of current digital collections, and 62% of those institutions used metadata assessment to evaluate content and scope. Of the institutions that did not assess collection content or scope, 33% cited lack of training as a barrier, 27% cited lack of personnel, and another 27% cited lack of documented standards or best practices.

In addition to content assessment, 65% of participating institutions collected usage statistics for their collections. When asked why they collect usage statistics, 19% of respondents indicated a desire to demonstrate the impact of collections on scholarship, 17% to improve digital collections services, and 17% to improve digitization selection decisions.

Only 29% of respondents collected reuse data. Collected data most frequently included citations in academic scholarship, published or reposted digital objects in digital media such as websites and digital exhibits, and sharing of digital objects through social media. Of the institutions that did not collect reuse data, 39% selected lack of documented standards or best practices as the main barrier, 26% selected lack of training, and 22% selected lack of personnel.

Implementation

Survey results suggest that the project team should provide training and shared best practices and principles to flexibly guide LDL participating institutions, which use a wide variety of processes for making digitization selection decisions and for assessing collection content, use, and reuse. Any attempt to superficially impose a set of one-size-fits-all standards or workflows would unnecessarily alienate some of our institutions. Our diverse community of institutions is better served by an ongoing dialogue around selection and assessment that leverages the expertise distributed across the LDL.

In an effort to begin building this dialogue, the project team facilitated the “LDL as Data Online Speaker Series”, during which we virtually hosted digital library practitioners and users from across the country. We recruited speakers who could address topics that respondents indicated would be of interest, such as selecting digital projects that fill representational gaps, and assessing users and reuse. These talks highlighted priorities such as inclusivity and community engagement, bringing them freshly to the forefront of community discussion.

We also hosted a content administrator focus group, during which we received more specific feedback about technical developments that would facilitate user-centered digitization selection and assessment.

Finally, the project team established the “LDL as Data Fellowship”, which supported four researchers in creating small-scale digital projects using LDL collection data. The fellowship concluded with the “Lunch with the LDL Fellows Brown Bag Series”, which allowed each fellow to present their project to LDL content administrators. This initiative not only produced specific LDL use cases, but also enabled discussion around ways in which our digital collections might be used and reused.

Outcome

After the conclusion of the “LDL as Data Online Speaker Series” and focus group, we distributed a second survey to evaluate the impact of these initiatives on strategies LDL administrators use to select content for digitization and to assess collection content, scope, use, and reuse. Fifty-seven percent of respondents indicated that the project enhanced their understanding of strategies for digitization selection, and 57% indicated that the project enhanced their understanding of assessing digital collections.

In addition to providing knowledge and best practices for selection and assessment, we hoped our efforts would foster stronger connections among LDL institutions. Seventy-one percent of respondents indicated that the project helped to build community among LDL contributors and administrators. Furthermore, when asked to explain how the project benefitted them, several respondents focused on the opportunity to exchange ideas with and learn from colleagues at other institutions. One respondent wrote,

Meeting with other LDL administrators (along with the knowledge gained from the LDL as Data Speaker Series) benefited me by providing an understanding of the various aspects of data and how we use that data....These meetings gave me a sense of what other LDL administrators were doing individually and as part of the LDL...Not only did I gain so much value from the collaborative meetings and the Speaker Series as an LDL administrator, but this helped me in assessing our digitization and data needs.

Another respondent wrote, “I can't stress enough the usefulness of the resources, communications, and collaborations. It has made me a better content administrator.”

There is still much room for improvement within the LDL as technology and as community. Although 38% of respondents report that they intended or hoped to make changes to how their institution decides what to digitize, and 57% reported that they intended or hoped to make changes to how they assess collections and use/reuse, lack of personnel and funding remain a major barrier to such developments.

Reflection

Given the multi-institutional and decentralized structure of the LDL, we find it ineffective to strictly implement standards that would apply to every contributing institution, from small cultural heritage museums to large academic libraries. Instead, we chose to take an evidence based approach to fostering community, productive dialogue, and flexible best practices.

Working with over 30 unique and autonomous institutions is never straightforward, but the evidence we acquired through our survey allowed us to facilitate discussion around topics relevant to our diverse community of administrators. Without this evidence, as project team members, we would have relied on our own experience to predict relevant topics, and likely would have excluded institutions that are dissimilar to our own. Based on what the project team learned, we continue to grow the LDL community through events, workshops, and other learning opportunities that bring us together in our shared pursuit of providing a unified portal to the digital cultural heritage of our state.

Author Contributions

Leah Duncan: Conceptualization, Funding acquisition, Investigation, Methodology, Writing – original draft Elizabeth Joan Kelly: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Writing – review & editing Sophia Ziegler: Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Writing – review & editing

References

Collections as data: Part to whole. (n.d.). Retrieved April 30, 2022, from https://collectionsasdata.github.io/part2whole/

The LDL as data. Louisiana Digital Library. (n.d.). Retrieved May 5, 2022, from https://louisianadigitallibrary.org/LDLasData

Ziegler, S., Powell Duncan, L, & Kelly, E. J. (2020). Louisiana digital library collections as data intro survey: Selected survey results to support reports and publications. LSU Digital Commons. https://digitalcommons.lsu.edu/libraries_pubs/97