Data science in data librarianship: Core competencies of a data librarian

Abstract

Currently, data are stored in an always-on condition, and can be globally accessed at any point, by any user. Data librarianship has its origins in the social sciences. In particular, the creation of data services and data archives, in the United Kingdom (Data Archives Services) and in the United States and Canada (Data Library Services), is a key factor for the emergence of data librarianship. The focus of data librarianship nowadays is on the creation of new library services. Data librarians are concerned with the proposition of services for data management and curation in academic libraries and other research organizations. The purpose of this paper is to understand how the complexity of the data can serve as the basis for identifying the technical skills required by data librarians. This essay is systematically divided, first introducing the concepts of data and research data in data librarianship, followed by an overview of data science as a theory, method, and technology to assess data. Next, the identification of the competencies and skills required by data scientists and data librarians are discussed. Our final remarks highlight that data librarians should understand that the complexity and novelty associated with data science praxis. Data science provides new methods and practices for data librarianship. A data librarian need not become a programmer, statistician, or database manager, but should be interested in learning about the languages and programming logic of computers, databases, and information retrieval tools. We believe that numerous kinds of scientific data research provide opportunities for a data librarian to engage with data science.

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

“…One of the main and overarching services, where libraries play a key role is RDM, which is a comprehensive set of activities for the organization, storage, access, and preservation of data [11]. It includes services, tools and infrastructure that support the management of research data across its lifecycle [6].…”

“…The former continue to follow many paths of traditional librarianship. On the other hand, they are bound to create new library services and focus on acting as facilitators in all stages of the scientific research [11]. Having varied educational and professional backgrounds, and being engaged in different types of work.…”

“…Data scientists are tied to business environments, which is characterised by shortterm tasks, reliance on statistical tools, and relational data modelling, which have not been preferred by academic libraries. Nonetheless, data sciences' methods and practices may turn out to be applicable to data librarianship, especially in regard to analytical and statistical services [40], data visualisation [41], programming languages, and their logic [11]. Familiarity with the disciplinary norms and standards of the given field is a must for both groups of professionals [6].…”

Quality of Open Research Data: Values, Convergences and Governance

“…The distinction between these two types of data is clear. Research data are the outputs of a systematic investigation that involves observation, experiments or the testing of a hypothesis [4], and consists of "heterogeneous objects and items used and contextualized, depending on the academic discipline of origin" [5] (p. 3). In contrast to this, data used in the corporate world are usually seen to be company assets, the latter being defined as a resource controlled by a business entity as a result of past events or a transaction, and from which "future economic benefits are expected to flow to the entity."…”

Data literacy in academia: Basics and pedagogical views

“…The nature of data literacy in academia As said above, this paper focuses on data literacy, applied in academic environments, therefore related to research data, which similarly to other types of data, needs to be managed. Research data management (RDM), consisting of a comprehensive set of activities for the organization, storage, access, and preservation of data (Semeler, Pinto, & Rozados, 2017) is inconceivable without data literacy training, because people, who will use research data need education about how to understand, interpret, and apply what they find, and researchers are no exception from this rule. Seen from a slightly different angle, data literacy instruction is often the first step in supporting researchers (Martin, 2014), then complemented by RDM, while we should not forget about data curation that may be compared to curating a museum collection for exhibit rather than for internal storage, extended by data preservation that involves frequent validation checks and backups (Thomas & Urban, 2018).…”