Big Data and Academic Libraries: The Quest for Informed Decision-Making

Abstract

Libraries remain one of the last places on campus where the purging of usage data is encouraged and "tracking" is a dirty word. While some libraries have demonstrated the usefulness of analytics, opponents bring up issues of privacy and debate the feasibility of student-generated library data for planning and assessment. Using a study conducted at the University Library, California State University, Long Beach, the authors of this article identified practical knowledge of data research that academic librarians will benefit from understanding. Readers will learn about the role campus culture plays in data gathering, be exposed to the complexities of learning analytics and Institutional Review Board (IRB) clearance, and read how the authors weighed the ethical use of big data analysis for assessing students.

Text mining applications to support health library practice: A case study on marijuana legalization Twitter analytics

“…Big data and data mining projects are gaining traction in many industries. Libraries use data‐driven assessment in a number of areas including ‘financial expenditures, such as serials, interlibrary loan, and online purchases, to other areas of library service’ (Travis & Ramirez, 2020, p. 34). While several data mining techniques exist, one interesting classification approach for libraries involves a framework that includes four quadrants: service analysis, usage analysis, quality analysis, and collection analysis (Siguenza‐Guzman et al, 2015).…”

“…One of the ways to assess library service is determining the relationship between library instruction and how library workshops may increase information literacy skills and student learning. This may be achieved by analysing course‐generated data and student grades (Travis & Ramirez, 2020). In another study, researchers used Google Analytics' Event Measurement to improve user experience on the library website (Vecchione et al, 2016).…”

Revamping the academic library use data capabilities: the Greek library science postgraduates’ perspective

“…Encouraging reports on the international record, especially the Anglo-American and Australian context, keep piling in since 2008 (Travis and Ramirez, 2020; Showers, 2014; Perry et al , 2018), detailing new tools, models and methods of library use data integration in institutional data systems and reporting statistically significant positive implications for the communities they represent. Conversely, universities across Europe are still hesitant to the perspective of linking the library to wider campus student information databases.…”

Exploring current practices and Greek and Spanish academic library stakeholder perceptions towards reconceptualizing in-library use data collection ecosystem

“…These models are envisioned to incorporate both retrospective and forward-acting aspects but also encompass the bureaucratic and economic system in which they reside, respond to fundamental user requirements and support at the same time long-range strategic objectives and near-term tactical needs. Their successful implementation depends upon (1) the development of an understanding of the technology in terms of its advantages, its costs and the relation to processes already in use, (2) feasibility proof assuring that early roadblocks such as the lack of librarian data analytics expertise, interoperability issues, inconsistent data collection methods, library culture and external pressures do not become permanent barriers, (3) collaborations involving multiple organizations and providing the base of support for early-stage innovation and finally (4) availability of both adequate capital in the form of grants and project funds and ''knowlegeable team members'' to support data collection and analysis in a continuous assessment cycle (Travis & Ramirez, 2020;p. 41) Library science bibliography since 2008 is abundant with publications and studies reporting progress on the exploration of whether and how library service point data integration with campus information systems can improve the accuracy of a learner profile, the adaptation, and personalization of content, inform thoughtful and well-planned manipulation of space (McKinstry, 2004), and information literacy instruction delivery, increase student engagement and belonging in higher education (Thomas, 2012) and keep libraries relevant and performing at their best.…”

“…Concerns raised around privacy intrusion or dataveillance-user tracking without their consent-that can do more harm than good according to Jones andSalo (2018 in Travis &Ramirez, 2020), student profiling (Sclater, 2015;, Asher et al, 2018;Tsai et al, 2018) deprivation of student agency, the dangers of reaching oversimplified conclusions about patron behavior and the amateurization of big data analysis (Chen et al, 2015), the insufficient proof of return on analytics investments are sustaining faculty, especially humanists and older educators' skepticism on the usefulness of analytics tools to move toward a data-driven approach to decision making on their campus (Yanosky & Arroway, 2015;Blankstein & Wolff-Eisenberg, 2019). Moreover, the complexity involving collaboration between academics and library staff (Pham & Tanner, 2015) has also contributed to keeping analytics uptake to low numbers on both an institutional and library level.…”