Curation checklist for data packages in Yoda@WUR

Curation checklist for data packages in Yoda@WUR This document provides a checklist for data managers to review data packages as part of the curation process for Yoda@WUR. Detailed curation guidance can be found here. Step 1 – Overview of the data package. First take a quick look at the metadata, documentation and files to get an impression of the context of the data package before reviewing it in depth. Step 2 – Reviewing the Yoda metadata form. Fields marked in red* are mandatory. Field Description Title* A descriptive title for the data package. If the data is part of a publication, consider “Data underlying the publication: [title]”. Description* An abstract that provides an overview of the data package. Discipline* Disciplines and subdisciplines that are relevant to the data package. Version Version is either a number or a date. Language of the data* The language of the data package. Default is English. Collection process The start and end dates of the data collection period. Location(s) covered The specific geographic location, when applicable. Period covered The time period covered by the data. Keywords* At least 4 relevant key words to describe the data package. Only 1 keyword per field (use + for more keywords). Related resource Any resources (articles, data packages, software) related to the data package, including title, relation type, and persistent identifier (e.g., DOI/URL). Retention period Default is 10 years. Any deviations from this should be explained in the ‘retention information’ field. Retention information Text field for remarks about the retention period. Embargo end date Specifies the date the embargo ends, and the data package becomes available. Note: the metadata will already be published. Data type A relevant data type is selected. Data classification* If a value other than ‘Public’ is selected, first contact a PO or ISO for consultation before publishing. Name of collection A collection that is consistent across data packages. Funding reference The funding source(s) of the data package, including the funding organisation and award number, when applicable. Remarks Additional information, context, or comments about the data package. Creator* The author(s)/creator(s) of the data package, including their affiliation and person identifier (e.g., ORCID, Scopus ID). Contributor Additional people who contributed to the data package, including their type of contribution, affiliation, and person identifier. Data package access* The access level, i.e. open, restricted, or closed, that matches the data classification and licence. Licence* The chosen licence for the data package. When custom is selected, a licence.txt file accompanies the data package. Step 3 – Reviewing the documentation. README.txt file: the README file is a document that accompanies the dataset, containing essential metadata and providing a comprehensive explanation of the data package that makes it understandable and reusable. The README file offers additional context and instructions that surpass the structured information typically found in the metadata form. More information on data documentation can be found here. Some minimum elements of the README file are: ❑ Title of data package ❑ Author(s)/Creator(s) and affiliations ❑ Rightsholder(s) to the data and/or software ❑ Description ❑ Folder structure and contents / file formats ❑ Methods ❑ Materials and software (requirements, installation, instructions) ❑ Explanation of the variables or reference to a codebook ❑ Licence Codebook: a codebook is a document that defines each variable and term in the data package, including their names, labels, and types. Though it is optional, a codebook can be helpful for organising, understanding, and interpreting the contents of the data package. Step 4 – Scanning the data files. 1. Begin by opening (some of) the data files to verify their completeness and functionality, e.g. if they can be accessed and read without any issues. 2. Ensure that all expected files are present, look for any missing files or errors that might indicate problems with the data package. 3. Examine if the folder structure is clear and logical, confirm that the files are appropriately named and are in preferred file formats. 4. Open some of the files to check the dataset for specific types of data, this includes: o Personal data (identifying information), such as names, addresses, postal codes, photographs, phone numbers, WIFI-network ID, GPS locations, personal information revealing a person's ethnic origin, political opinions, religious beliefs, sexual orientation, and genetic or biometric data. o Data for which WUR does not possess the legal rights or ownership, such as data from a company or another research institute. If any of the files contain personal data, otherwise sensitive data, or data with rights issues, please contact [email protected] for further assistance.