Research Data and Reproducibility

A Data Management and Sharing Plan (abbreviated as DMSP or DMS Plan) is a required component of many grant applications. Different funding agencies have different requirements for data management and sharing, but in general, they typically describe:

• The data collected or generated during the research project
• How the data will be documented
• How the data will be organized, secured, and stored during the project
• How the data will be shared
• Where and how the data will be preserved and archived for long-term access
• The roles and responsibilities for managing the data

We have specific information regarding NIH's Data Sharing Policy.

The types of data that may be subject to a data management and sharing policy may include:

• Primary data
• Datasets that have been cleaned and processed for final analysis
• Data that supports findings and figures in publications
• Data required to reproduce or replicate published findings or figures
• Computational artefacts such as models, algorithms, scripts, code, or software
• Metadata that describes all the above

Implementing good data management practices makes data sharing easier.

Data management and sharing policies generally do not require you to share:

• Legally or ethically restricted data such as personally identifiable information that cannot be properly de-identified, data subject to privacy laws (e.g., HIPPA or FERPA), or data involving vulnerable populations where sharing could cause harm.
• Internal or preparatory materials such as lab notebooks, preliminary analyses, drafts of manuscripts or plans for future research.
• Physical objects such as lab specimens
• Data not needed to support published findings

While sharing these types of data is not required by data management and sharing policies, open and reproducible science practices increasingly encourage sharing protocols and raw data.