In October 2020, the Metadata Research Center and Drexel College of Computing & Informatics — in partnership with UC San Diego Library, Montana State University Library and Online Computer Library Center — was awarded nearly $900,000 from the Institute of Museum and Library Services (IMLS). The funding launched the nationwide Library Information Science Education and Data Science-Integrated Network Group (LEADING). As part of this initiative, UC San Diego is serving as a home for data science fellows as well as a hub for fellows to concentrate on big data and data mining.
LEADING is a three-year project that provides fellowships for graduate students in library and information science and early- to mid-career librarians. Each year, the Library will host multiple fellows who will implement newly acquired data science skills through applied projects. In the first year, three fellows focused on two projects — applying data science techniques to improve metadata quality in the Farmworker Movement Documentation Archive, and evaluating the factors that influence building a community of practice in information and data science.
[ FARMWORKER MOVEMENT DOCUMENTATION ARCHIVE ]
How can data science be applied to community-created collections of historical material? Fellows Ateanna Uriri (right), metadata librarian at University of Texas Rio Grande Valley, and Hiva Kadivar (center), assistant to the Middle East Studies Librarian at University of North Carolina at Chapel Hill, worked with mentors at the Library to search for the answer.
UC San Diego’s Data Science Librarian Stephanie Labou; Assistant Program Director for Scholarship Tools and Methods Annelise Sklar; and Associate University Librarian for Scholarly Resources and Services Roger Smith provided the fellows access to the Library’s Farmworker Movement Documentation Archive, a rich digital archive documenting the United Farm Workers (UFW) Movement in Central California from 1962 to 1993. The archive, which was developed by LeRoy Chatfield and acquired by the Library in 2014, includes a wide variety of information on the activities, accomplishments, challenges and work of Cesar Chavez and the farmworkers who participated in the movement.
Content from this archive is regularly used in classrooms and cited in scholarship, news articles and online exhibits. However, because it retains its original organizational structure and coding, it is difficult to grasp the extent of the history and connections between people involved in the movement without reading every document and viewing every photo.
The fellows were tasked with identifying ways to make the site more accessible for research and learning purposes through the strategic application of data science approaches, such as text mining, web scraping, topic modeling and interactive visualization.
Ultimately, this effort would yield a template for how “legacy” digital collections can be reimagined using efficient and transformative methods and tools.
[ EVALUATING A COMMUNITY OF PRACTICE ]
In this study, UC San Diego’s General Instruction Coordinator Librarian Crystal Goldman (far left) and University Librarian Erik Mitchell (the “researchers”) used a survey paired with interviews to examine participants’ engagement with and their perceived value of a community of practice (CoP) created to build data science expertise in the library and information science field.
Goldman and Mitchell used a social learning theory-based conceptual framework for assessing the value created in CoPs. The main form of data collected as part of this framework are “value creation stories,” which include five specific types of value: immediate value, potential value, applied value, realized value and transformative value. The framework also stresses the importance of “learning loops” that provide feedback to community members about how what was learned in the CoP worked or did not work in individual practice, thus creating further opportunities for social learning and value creation.
For the first phase of this study, the researchers created an online survey designed to collect value creation stories, demographic data and experiences with diversity, equity and inclusion in the CoP. For the second phase, the researchers conducted interviews with survey respondents who were willing to participate in a follow up interview about their experiences as CoP members. With both phases of the study complete, Goldman and Mitchell are now in the process of analyzing the data and compiling their findings into a scholarly journal article.
In the second year of the LEADING Program (beginning July 2022), new fellows will be engaged in projects focused on data publishing as well as continued work on the development of CoP in data and information science.
“It is my hope that this program builds on the commitment our Library has for mentoring students and early career professionals. I am excited to see our fellows’ work and have our Library learn from their expertise. We would like to thank the IMLS for their generous funding to support this program,” said Mitchell.
To learn more about the LEADING program and fellowship projects, visit cci.drexel.edu/mrc/leading.
—
Credit: The image embedded in the body copy above was captured by Jon Lewis and was pulled from the Library’s Farmworker Movement Documentation Archive. Fellows Ateanna Uriri and Hiva Kadivar were tasked with identifying ways to make the site more accessible for research and learning purposes.