Data Sharing Checklist

Overview

Exercises: 35 min

Questions

• Why are we sharing the data?

• Which data should we share?

• How much time/money will it cost?

Objectives

• Understand how the key data sharing questions relate to your work

Digital Curation Centre

The DCC has a wealth of expert guidance on data sharing and curation. They are responsible for the widely-used DMPonline data management plan form.

In this lesson, we will discuss a condensed form of the DCC guide on deciding what data to keep.

What could the data be used for?

• Reproducibility: ensuring your data are consistent with the reported results.
• Reanalysis/Meta-analysis: answering different questions to the ones already asked; answering the same question analysed in a different way; contributing to a wider understanding through integration with other similar data.
• Common example: acting as a reference point to facilitate discussion and shared understandings; contributing to a body of observed facts that theories must account for.
• Teaching: allowing learners to perform analysis on real data.
• Personal backup: preserving access across computers and institutions.

Discussion 10 min

Which of the reasons above are relevant to your research?

Are there other reasons you might share data?

What policies apply?

• Journal policies
• Institutional policies
• Funder policies
• Government policies, which may include security classification and freedom of information policies.
• Contract/patent rules
• Use for legal/public/police enquiries
• Personal data rules (GDPR)
• Ethics and consent

Discussion 10 min

Which of the policies and considerations above are relevant to your research?

Does your institution have a data sharing policy?

Can you share anonymised data gathered from people without express consent?

Do other policies apply to your research not mentioned here?

What data should be shared?

• Relevance: are the data relevant to reported results?
• Quality: is the data sufficiently complete, coherent, valid, etc. to be useful?
• Metadata: can we provide enough information about what the data are, how they were collected, etc?
• Demand/appeal: will the data be of use to specific people? Will the data contribute to existing resources that are of use? Do the data support published research?
• Uniqueness: can the data be easily reproduced (e.g. random numbers from a seeded generator)

Discussion 10 min

How do these considerations relate to your research data?

Are there other considerations that apply to your data?

Costs of sharing

• Curation: getting the data in an open format; adding metadata and creating a data dictionary.
• Storage: housing in a repository and keeping available.

Discussion 5 min

Who will pay these costs?

Are they time/money costs?

Will they pay off in the long run?

Might they benefit you even if you didn’t share?

Are there other costs that apply to sharing your data?

Key Points

• There are several common reasons for sharing data, including verification of results and further analysis.

• Generally, if you can share data, do.

• Sharing data has time and money costs - who is paying them?