Data integration as the major mode of data reuse

Abstract

Data sharing has been emphasized in scientific research in the era of open science; however, the concept of data sharing is ambiguous, and the real situation of data sharing is unclear. One reason is that most studies focused on researchers' recognition, motivation, and other contributing factors and did not thoroughly discuss the actual practice of data sharing for preparing research articles. To explicitly understand the current status of data sharing, the authors considered and investigated primary reuse, data integration, and data release distinctively as the actual practices of data sharing. A total of 237 articles published in 2018 and registered in the Web of Science global citation database were manually checked. Unlike the scarcity of data reuse (8.0%) or data release (5.1%), existing data were most frequently reused in the form of data integration (20.7%). The result implies that further studies on detailed data reuse are required to clarify the concept of data sharing.

MetaFAIR: A Metadata Application Profile for Managing Research Data

“…Furthermore, the inclusion of elements for describing relationships and dependencies among data objects and workflows reflects the importance of this type of information for ensuring that researchers can discover dataset by employing attributes specific to their domain of research and reuse various kinds of data objects with a good understanding of the reusability of datasets. In this way, the MetaFAIR AP seeks to support both the primary reuse and data integration described by Sakai et al (2020) through more effective data release.…”

“…Data integration is another method of data reuse that integrates existing data with original data generated for a study. Data release is the process of making original data independently available for future reuse (Sakai et al, 2020, p. 1). These definitions depict common methods by which researchers work with data in their research.…”

“…Thus, both data integration and release are forms of data reuse. Based on the work of the University of Pittsburgh Library System (2020) and Sakai et al (2020), data sharing and reuse can be understood to be the two ends of the continuous cycle of scientific research. In the past five years, the release of the FAIR Data Principles has caused data sharing and reuse to shift from being latent concepts to being a focus of RDM scholarship.…”

A Bibliometric Overview of Open Science Research

“…Reproducibility is important for statistical analyses [42] and indicated research quality [16]. Interoperability regards the integration of two or more datasets into a whole [43] promoting re-usage [44]. Making data FAIR involves meta-data definition and management [45], which is necessary for multidisciplinary research works [46].…”