Open Science
Open science is a movement to make the data collected through scientific research freely available to all. Federal policies now require an open science approach, mandating public access to federally funded research data. Despite the clear benefits of open science, researchers face many challenges in sharing data, including the need for data platforms and concerns about data privacy, confidentiality, and intellectual property rights. Researchers may also worry about attribution for their data and interpretation of the data by other researchers.
Developing new incentives for researchers may be the most important strategy to promote data sharing. Researchers now have many reasons to control their own data until they have gotten the maximum publication value from it, and few incentives to share the data. Career advancement depends more on publishing scientific papers than on creating the datasets those papers are based on. New incentives could credit and reward data sharing in ways that encourage open science.
Data governance and technical initiatives are also needed for an integrated approach to data sharing. Implementation of technical standards, data platforms, and tools for interoperability will all help promote data sharing. So will new approaches to data governance that manage research data from different sources in an integrated way. Some potential best practices include:
Require data sharing and publication as a condition of research funding and help researchers meet that requirement. While federal guidelines now include an expectation of data sharing for federally funded research projects, the guidelines are not yet very specific. Guidelines now require grantees to develop data management plans with an expectation that, at minimum, the data underlying publications will be made accessible and shared. Federal funders could tie grants to clearer, binding requirements to adopt open standards and share data publicly to the greatest extent possible, taking privacy and other concerns into consideration.
At the same time, federal grant-makers can provide positive incentives and help researchers meet the data-sharing requirement. They can value open data more highly in funding decisions, giving extra points to grant applicants who are committed to sharing their data. Funders can also provide sample data management plans for federal grantees. While grant applicants are now required to develop these plans, it would be beneficial to clearly encourage data sharing within the goals and recommendations for data management and data infrastructure development.
Use new incentives to promote research data sharing more widely. Currently, there are many incentives against sharing research data and few that support it. The challenges include both cultural and pragmatic obstacles. The current scientific culture is not to share data, but for individual researchers to hold datasets for their own use. The academic model does not reward data sharing. Since academics are rewarded for publishing peer-reviewed articles more than for publishing datasets, researchers want to get maximum publication value out of their data before releasing it. They may also worry about attribution for their data and interpretation of the data by other researchers. In addition, data sharing can be expensive, and it is not clear how to fund it.
New ways to reward data-sharing through funding, tenure decisions, and other career incentives could significantly increase data-sharing by researchers. A key is to ensure that researchers receive systematic and meaningful credit for sharing their data. Data citation systems, similar to the citations for published papers, could help researchers gain credit for their work, measure the impact of their research, and advance professionally. They could form the basis for “report cards” that researchers can access to see how their data is being used. This would be similar to the way some organizations now support the use of open source software.
While focused on researchers, improved citation systems for data could also help federal agencies and research institutions track the use and impact of the data they produce. This effort could be supported by guidance from the U.S. General Services Administration and the Office of Management and Budget to federal agencies, in partnership with organizations that support open science.
Develop collaborations and outreach to collect, manage, and publish data. Scientists in many research domains have begun to form wide-ranging collaborations around data. These collaborations make it possible for researchers to draw on each others’ work to accelerate the pace of science. They also enable scientists to draw on diverse groups for data collection: for example, patients who volunteer data in health studies, and the population at large for a number of citizen science projects. Prominent examples of data collaborations to support scientific research include the Study of Environmental Arctic Change (SEARCH), the BRAIN (Brain Research through Advancing Innovative Neurotechnologies) Initiative, and the National Institutes of Health (NIH) Commons.
Featured Examples
