LibGuides: Data Services: Sharing Data

Why Share Your Data?	Best Practices	Library Support	Funder Requirements
Journal and Publisher Policies	FAIR	Common Data Elements	Data Journals

Why Share Your Data?

Data Sharing has multiple benefits which include increasing the visibility of your research, facilitate new discoveries by other researchers, and meet funding requirements.

Increase your visibility by including your data in open access repositories. Raise your prominence by contributing to the field and keep your research relevant.

Facilitate new discoveries by allowing other researches to use your data sets. Data sets can be re-analyzed to answer a research question and prevent duplication of effort.

Meet funding requirements as many federal grants require data sharing around the time of publication. Individual journals and publishers also are requiring or encouraging data sharing in their editorial policies.

Make public assets available to the public through publishing data papers, research articles with your conclusions and submitting datasets to open repositories.

Other Incentives:

Data repositories create DOIs allowing for citations of the data
Data journals allow for data publications and full text articles
NIH Biosketch allows for the citing of non-traditional research outputs such as datasets
Data papers allow researchers to receive citations for a dataset like a publication
Sharing data can result in increased citations for published studies
Maximize transparency, accountability, and scrutiny of research findings

Open Data
An introduction to the legal, social and technical aspects of open data.
Retraction Watch
Blog about research articles being retracted for not being reproducible, lost data, etc.
The State of Open Data Report 2019
The longest running longitudinal study on the subject, to examine attitudes and experiences of researchers working with open data – sharing it, reusing it, and redistributing it.

Not all data can be shared freely.

Restrictions on

Threatened or endangered species
National security and classified research
Export controls (technologies)
Personal Health Information
Data involving children or prisoners

Best Practices

Ensure you can share your data easily:

Use Open Source file formats (ASCII)
Provide software information in metadata (software name, version, operating system)
List number of files in file structure in metadata
Select a consistent file format
Select files that can be edited regardless of application
Files should follow a documented standard
File should stay unencrypted and uncompressed
Upload datasets to open repositories for easy access
Keep it clean by avoiding missing values or unclear column headers
Understand funder/institution policies on data sharing
Remove personal data

Funder Requirements

Funder	Policy Effective	Publication Requirements	DMP Required?	Data Requirements
AHRQ	October 1, 2015	Provide full public access to publications no later than 12 months after the official date of publication. AHRQ will archive it's publications in PMC	Yes	Small data sets to be released quickly, larger data sets are to be released in waves as it becomes available or as main finders are published More Info
CDC	January 2015	Data available coincident with publication of paper	Yes	Data set released within 30 months after the end of data collection or generation, expect surveillance data which should be made available within a year More Info
FDA	December 29, 2015	Submit final peer-reviewed manuscript to NIHMS upon publication; made available in PubMed Central within 12 months	Yes	Share data underlying research papers when the paper is published More Info
IES	2012	Publications submitted to ERIC within 12 months of publisher's official date of final publication	Yes	Final Research Data and free of personal identifiers; data available no later than time of publication More Info
NIAID	July 2017	Submit final peer-reviewed manuscript to NIHMS upon publication; made available in PubMed Central within 12 months	Yes	A data sharing plan is required only for applications requesting $500,000 or more. More Info
NIH	April 7, 2008	Submit final peer-reviewed manuscript to NIHMS upon publication; made available in PubMed Central within 12 months	Yes	For grants over $500,000 a data sharing plan must be included in the application. Key Elements
NSF	January 18, 2011	Submit final peer-reviewed manuscript to NSF Public Access Repository upon publication; made available within 12 months in repository and/or on participating publishers’ sites	Yes	Investigators are expected to share with other researchers the primary data, samples, physical collections and other supporting materials created or gathered in work under NSF grants More Info

Find Additional Federal Grants

NIH Sharing Policies and Related Guidance on NIH-Funded Research Resources
It is NIH policy that the results and accomplishments of the activities that it funds should be made available to the public. PIs and funding recipient institutions are expected to make the results and accomplishments of their activities available to the research community and to the public at large. The following links highlight selected NIH policies and related guidance on sharing of research resources developed with NIH funding.
NIH Frequently Asked Questions on Data Sharing

Journal and Publisher Policies

Journal or Publisher Name	Policy Summary
Annals of Internal Medicine	Manuscripts submitted to Annals that report the results of clinical trials must contain a data sharing statement that meet the ICMJE recommendations.
BMJ	BMJ requires a data sharing statement for all research papers. For papers that do not report a trial, we do not require that the authors agree to share the data, just that they say whether they will.
ICMJE	As of 1 July 2018 manuscripts submitted to ICMJE journals that report the results of clinical trials must contain a data sharing statement.
JAMA	For reports of randomized clinical trials, authors are required to provide a Data Sharing Statement to indicate if data will be shared or not.
The Lancet	From July 1, 2018, all submitted reports of clinical trials must contain a data sharing statement, to be included at the end of the manuscript.
New England Journal of Medicine	The ICMJE and, therefore, NEJM require investigators to submit a data-sharing statement (2018) and register a data-sharing plan when registering a trial (2019).
PLoS	The data underlying the findings of research published in PLOS journals must be made publicly available.
PNAS	To allow others to replicate and build on work published in PNAS, authors must make materials, data, and associated protocols, including code and scripts, available to readers.
Science	The Science Journals support the efforts of databases that aggregate published data for the use of the scientific community. Therefore, before publication, large data sets must be deposited in an approved database and an accession number or a specific access address must be included in the published paper.
Springer Nature	At Springer Nature we want to enable all of our authors and journals to publish the best research, which includes achieving community best practices in the sharing and archiving of research data.
Sherpa Juliet	Sherpa Juliet enables researchers and librarians to see funders' conditions for open access publication.

FAIR

Findable- Data and metadata are easy to find by both humans and computers.
Accessible- Data use is open to the greatest extent, allowing others to query or copy data for their own use.
Interoperable- Data can be interpreted by a computer easily.
Re-useable- Data and metadata is well described for both humans and computers to replicate or combine data with other datasets.

More Info

FAIR Data Principles explained

FAIR Guiding Principles

Data Stewardship Wizard
Create Smart Data Management Plans for FAIR Open Science
FAIR Data tools
These tools assist in "fairificiation" of data.
Fair Sharing
A curated, informative and educational resource on data and metadata standards, inter-related to databases and data policies.

Common Data Elements

Common Data Element (CDE) - A data element that is common to multiple data sets across different studies.

Certain types of CDEs are sometimes described:

Universal - CDEs that may be used in studies, regardless of the specific disease or condition of interest, e.g., demographic information of study subjects or medical history
Domain-specific - CDEs that are designed and intended for use in studies of a particular topic, disease or condition, body system, or other classification, e.g., Parkinson's disease, Alzheimer's disease, diabetes, ophthalmology. Some domains are broadly applicable to a wide range of studies, while others are more useful in specific fields of clinical research.
Required - CDEs that are required or expected, as a matter of institutional policy (e.g., research funder or performer), to be collected for all subjects in studies of a particular type, e.g., NIH-funded studies of neurological disease
Core - CDEs that are required or expected to be collected in particular classes of studies, e.g., any study of neurological disease or cancer, any genome-wide association study. Other, domain-specific common data elements may be suggested, expected, or required for collection, depending on the more specific focus of the study (e.g., Alzheimer's disease or ovarian cancer)

NIH Common Data Element Portal

NIH CDE Repository

Draft of U.S Core Data for Interoperability of Clinical Research

Data Journals

Data papers are an easy way to share data. They are not traditional research articles, as their purpose is to simply describe your dataset.

There are over 200 data journals, with 120 of those in biological, medical or health care fields. They are mostly open access, indexed in widely used databases, peer reviewed, and some have impact factors.

The benefit to submitting a data paper is to double the publication output from a single research project. Publishers consider data papers complementary to research articles, and not a prior publication.

The paper is easy to write as it is usually brief and you are provided an article template to follow.

Since data papers can be cited, it is easy to track reuse of your data and contribute to one's h-index.