Why share your research data?
Many biomedical and population health journals and funders now require researchers to make their de-identified research data publicly available when they publish their results. Journals like PLOS, PNAS, Science, and Nature require de-identified data sharing as a prerequisite to publication. Funders like Howard Hughes, NIH, and the Gates Foundation now require that grantees make data public alongside their articles, and starting in 2023 NIH will require a data sharing plan from all grantees. But what does it mean to share your data? This page will give you an overview of the process.
UCSF Guidance for Data Sharing
UCSF now has guidance for sharing de-identified data to meet publisher and journal requirements. Check out the step-by-step guidance to learn how to plan for data sharing, and share data in accordance with UCSF policies.
Select data and documentation for sharing
What data will you share? This is largely dependent on your field of research, but you should consider what someone else in your field would need to validate your results
What documentation needs to accompany the data? Data by itself is seldom useful. What other dictionaries, metadata, code would someone need to use your data?
- PLOS One Data DefinitionPLOS guidance for authors on what data they should be sharing
- ICPSR Guide to Social Science Data Preparation and ArchivingAn excellent overview of social science data collection, analysis, and archiving including how to properly document research data and how to prepare data for sharing
- NIH Guide to Genomic Data SharingStep-by-step instructions to submit genomic data to NIH and NCI repositories.
Ensure you have permission to share
Do you have consent to share?
If your research involves human subjects did you mention data sharing in your IRB and informed consent documents?
- UCSF IRB Consent FormsTemplates and sample consent forms for human subjects research now updated with language about data sharing.
- ICPSR Recommended Informed Consent Language for Data SharingSample text to use in consent forms from the ICPSR data repository
De-identify your data
Has your data been de-identified?
It is important to properly de-identify your data to reduce the risk of identifying individuals in a dataset. UCSF now offers a service to validate this.
- UCSF De-identification ResourcesOverview page of UCSF de-identification services and resources.
- Preparing Raw Clinical Data For PublicationMany peer reviewed journals now require authors to be prepared to share their raw, unprocessed data with other scientists or state the availability of raw data in published articles, but little information on how such data should be prepared for sharing has emerged. Iain Hrynaszkiewicz and colleagues propose a minimum standard for de-identifying datasets to ensure patient privacy when sharing clinical research data
- Guide for De-identifying Qualitative ResearchThis guide from the Qualitative Data Repository discusses best practices for dealing with direct and indirect identifiers in research