Collection development or data-driven content curation? An exploratory project in Manchester

Abstract

Purpose – Collection development in a post-subject librarian age needs to be done differently; utilising data, metadata, analytical tools and automation more fully may offer new possibilities. The purpose of this paper is to report and evaluate an exploratory project into new techniques for collection development at the University of Manchester Library. Design/methodology/approach – The project employed a cross-team approach where a relatively large number of staff tried some innovative and experimental approaches to individual aspects of a large and complex task in a large, research-intensive university library. The overriding aim was to exploit data to support decision making and to push automation as far as possible. Findings – The quality of (meta)data remains a huge hindrance to data-driven approaches. A proper understanding of usage data is an urgent but intractable issue. Human input and relationships are still important. Data are nothing without analysis, and many librarians currently lack the data fluency to work confidently in a world of dynamic content curation. Practical implications – Librarians need both to re-skill and to change their self-identification and the philosophy that underlies it if they are to achieve confident, data fluency. Originality/value – The University of Manchester Library was one of the first libraries in the UK to make a thoroughgoing structural change from subject-based to functional teams. This paper will be of value to other libraries moving in this direction, and to those looking to make more use of data-driven decision making in collections management.

Research Management Systems: Systematic Mapping of Literature (2007-2017)

“…From the studies analyzed, Table VII and Figure 6 show that the most frequent type of contribution was Process, with 36%, followed by Model with 27%, Framework, and Classification with 12% each, Mapping with 6%, and finally Method and Tool with 3%, respectively. This type of contributions coincides with the reported studies in the literature, of note being: Process [14], [15]; Model [16]; Framework [18]; Classification and Mapping [19], and Method and Tool [20]. One of the most representative articles on mapping in relation to the Process contribution states that a cluster analysis process was carried out on the bibliographic data of Slovenian researchers and the disciplines to which they belong, using their country's SICRIS, and from there it discussed the understanding of the structure of collaboration and cooperation in sciences and the research and development policies [15].…”

“…Research on CRIS has made significant contributions to strengthen such systems and to ensure their interoperability. Contributions have been made regarding (a) processes that had allowed institutions to increase and enhance their publication count compared to previous years, when their collections were handled with a production virtually invisible to the community of specialists [14], [15]; (b) modeling, with the introduction of new publication models [16], including open access with the help of CRIS, which made it possible to establish links between individual researchers, their publications and the funding obtained; (c) development of frameworks [17] for the management of information associated with research, with a proposal to use the rapid application development method to design reusable modules and integrate frameworks to ensure the security and integration of the data [18]; (d) mapping and classifications, with proposals made for the conceptualization of science based on collaboration and mapping of the researchers' competencies [19], this type of information has been obtained from the SICRIS (Slovenian Current Research Information System), Slovenia's database of national research projects; (e) methods and tools, such as the evaluation of an exploratory project on new techniques for the development of collections at the University of Manchester Library [20], using innovative methods to ensure the construction of collections for the future and to promote automation, proposing a prospective method: to use its CRIS for the development of its collections. Also, the use of Symplectic Elements© has also been reported as a tool to more efficiently measure the impact of their research.…”

Identifying new roles for academic libraries in supporting data-intensive research

“…Researchers, including teaching staff members and (doctoral) students do not need more data, but require having the right data [4]. However, data is of no use if not analysed [5].…”

Data Curation in Academic Libraries as Part of the Digital Revolution

“…Data curation activities should be guided by the understanding that researchers do not need more data, but the right data (Borgman, 2015) and that data is of no use if not is not analyzed. Accordingly, librarians have to acquire analytical skills and abilities (Kirkwood, 2016). They also should set up appropriate and innovative new services, recognized by several stakeholders, including research funders, research organizations, the communities of researchers, as well as academic and special libraries.…”