Randomised study of DataCite metadata completeness on dataset uptake and sharing ------------------------------------------------------------------------ Version & Date: v1.1 14-Jan-2019 STATEMENT OF COMPLIANCE ----------------------- This document is a protocol for study of effect within the Open Data environment of DataCite. The study will be conducted openly and with the consent of the data providers, and all parties involved. No human data will be used in this trial. PROTOCOL SYNOPSIS ----------------- - Title - Randomised study of DataCite metadata completeness on dataset uptake and sharing - Objectives – testing level of metadata completeness - Design – Simple Random Control, Observation of outcome - Outcomes – comparison of extensive vs minimal metadata on effect of re-use and sharing of data - Study Duration – up to 1yr - Interventions - none - Number of Subjects - 1093 datasets - Population – All individual datasets within a study of vascular plants in and around the Ruili Botantical Gardens. INTRODUCTION AND BACKGROUND --------------------------- Background: In the context of open data, and data sharing it is considered good practice to make metadata both openly available and widely distributed. The use of Digital Object Identifiers (DOI), provided by registration agencies such as DataCite provides an additional level of security to users by assuring that the metadata is not only public, but also of a particular minimal standard and can be found at a stable location (namely the DOI URL). There are several DOI providers worldwide, but this study focuses on the DataCite provider only as other DOI providers may require slightly different metadata. A full list of possible DOI registration agencies can be found here (https://www.doi.org/registration_agencies.html). GigaDB is a database primarily hosting scientific research data associated with articles published in the journal GigaScience, which is published as a partnership between BGI (formerly the Beijing Genomics Institute) and Oxford University Press (OUP). All GigaDB datasets are published with a DOI registered via DataCite. Research question: ------------------ Does the extent of the metadata provided to DataCite influence the uptake and reach of the dataset DOI? If so, in what way? In layman's terms, when releasing a dataset is it worth spending the time and effort to generate extensive rich DataCite metadata over the quick and easy minimal metadata? Rational for study: ------------------- The answers to the research question will provide evidence to make decisions on the level of effort made to embellish (or not) the minimal DataCite metadata with additional information. Study objectives ---------------- Compare reach and uptake of minimal metadata DOI datasets to enhanced metadata datasets over the 1 year time period. Results available in this preprint: https://osf.io/preprints/metaarxiv/xv7tk/