Memorial University Libraries

Contact: Alison Farrell, Research Data Management and Public Services Librarian

Last Updated: July 7, 2022

Purpose:

This document will help researchers follow best practices for depositing research data for sharing and publication in the Memorial University Dataverse Collection.

Expectations for Depositors:

Many files may be created over the course of a research project. An important step for depositors in preparing their data for deposit is the selection or appraisal of which files should be kept. In making these decisions, depositors should provide enough supporting documentation for other researchers to understand how data were created, how to reproduce methods and findings, and reuse the data files.

When making decisions of what documentation to include with your data, consider what someone (or your future self) would need to know to understand, evaluate, analyze, or replicate your data without having to ask you.

When possible, depositors are strongly encouraged to include a version of their raw data, in addition to any processed data used in published analyses and figures. If raw data contain any sensitive information, depositors should follow best practices for de-identification and ensure they have the proper permissions to share before depositing.

Standards for Deposit:

Before depositing in Memorial’s Dataverse collection, depositors should make sure their dataset(s) meet the following standards.

1. Use consistent and comprehensible file names and file structures.

Following proper file naming conventions makes it easier to navigate and find specific files, and allows other researchers to understand and reuse your dataset.

•  Name files consistently
•  Keep file names short (< 25 characters) but meaningful
•  Do not use spaces to delimit words. Use capital letters, hyphens, or underscores
•  Do not use non-alphanumeric characters
•  Denote dates using ISO8601 standard YYYY-MM-DD (e.g. 2019-01-10).

2. Deposit your files in preferred file formats to support preservation and reuse.

The use of preferred file formats is important to support the long-term preservation of your research data. Consult the following resources for a non-exhaustive list of preferred file formats.

• UK Data Archive. Recommended formats for data sharing, reuse and preservation.
• DataverseNO. Prepare your data: Preferred file formats.

If appropriate, files may be deposited in their original file format, in addition to a preferred format.

If you have any questions about preferred file formats for your research data, contact Alison Farrell, Research Data Management and Public Services Librarian

3. Describe your dataset with rich metadata to make it easily findable.

Metadata, or data about your data, is an important element of datasets. Metadata answers questions about your data, such as:

• who created it?
• when and where it was collected?
• what is the file format of your data?

Metadata allows users to find your datasets and understand how the data was collected, organized, and used. Depositors must complete all required fields in the descriptive metadata. Depositors are strongly encouraged to complete geospatial metadata fields and subject-specific metadata fields, as appropriate. Consult the following resource for guidance on Dataverse metadata fields.

• Dataverse North Metadata Best Practices Guide

Memorial Libraries may suggest changes to the descriptive metadata for the purposes of discovery, reuse, and preservation.

For assistance with metadata, please contact librarian, Heather Pretty.

4. Include a ReadMe file to support correct interpretation and reuse of your dataset.

For research data to be read and interpreted correctly, it requires sufficient documentation. It is recommended that deposited datasets include a “ReadMe” file that includes the following information:

• Details about dataset creation
• Description of files contained in the dataset
• Information about dataset completeness
• Limitations on reuse

ReadMe files should conform to the following:

• ReadMe files should be saved as a Unicode UTF-8 plain text file (.txt). Alternatively, ReadMe files may be saved in PDF/A format.
• ReadMe files should use forced numbering in the filename (e.g. 00_ReadMe.txt) to make it appear at the top of the file overview.

Consult the following resource for a basic ReadMe file template:

• Cornell University. Readme template.

Expectations for Curators:

A data curator at Memorial Libraries will review all deposited datasets for alignment with the Deposit Guidelines. Depositors may be asked to make necessary modifications. Once a data curator has approved the submission, the depositor will be notified and the dataset will be published. All published datasets will receive a Digital Object Identifier (DOI) to allow the dataset to be cited.

Please note, Memorial Libraries does not attempt to judge the scholarly quality of deposited datasets, and trusts the judgement and research expertise of those who created and deposited the dataset. Thus, a determination of a dataset’s research quality is at the sole discretion of the contact person as named in descriptive metadata.

The requirements and guidance above have been adapted from:

University of Victoria. University of Victoria Dataverse Collection Deposit Guidelines.