Most research organizations’ incentive structures do not encourage (and may even inhibit) the broad sharing of research data. Universities’ promotion and tenure standards emphasize the publication of peer-reviewed articles in high-impact journals, and do not address the need for—or benefits of—publicly sharing the data behind those articles. This limits researchers’ willingness to share their data. Instead, they are incentivized to restrict access to important research data and maximize the number of publications they can derive from the data.
A 2014 survey of 90,000 recent authors of papers in health, life, physical, and social sciences, and humanities, showed just over half of researchers shared their data once an article was published. Of those who did not, 26 percent reported that they did not share data because they feared it would be scooped or misused. The same survey asked respondents what would motivate them to increase data sharing. Social science, humanities, and physical science researchers sought increased visibility and impact for their work; life science researchers sought guaranteed credit; and health science researchers sought guarantees regarding privacy and other ethical issues.
The research community could boost access to open research data by making changes to the current incentive structure, increasing visibility and ensuring credit for publishing data openly. However, it will take time and careful consideration to change the incentive systems that currently drive scientific research. To set a framework for this effort, the White House Office of Science and Technology Policy (OSTP) should work with major scientific journals, universities, and professional associations to undertake a 100-day scoping exercise on changing incentives for data publication. This exercise should identify:
• Promising strategies, including changes to the citation system, the promotion and tenure process, and other approaches;
• Key issues, including differences between scientific domains that impact data publication, economic considerations, and consistency of academic policies;
• Major partners to implement change after the scoping phase, including not only universities and journals but professional associations, such as the Research Data Alliance and CODATA, that address incentive issues through working groups; and
• Near-term actions that can shift incentives for data publication.
This scoping exercise should support changes in the scientific publication process. For example, journals can develop appropriate review systems to ensure that published datasets are rigorously reviewed and useful for additional research efforts, including replication of existing published work. Scientific publishers may also develop citation systems that give researchers visible credit when other scientists use their data.
OSTP could also explore requiring that all federal research funding agencies develop policies that require grantees to share the data underlying scientific publications for free at the time of publication.
In addition, OSTP could work in partnership with major American research universities to support new academic incentive structures for publishing reusable data. The incentives may already be starting to align naturally, as research shows that studies with publicly available datasets receive a higher number of citations than similar studies without available data. By incorporating open publication of datasets into promotion and tenure decisions for faculty, as well as tying open data to advancement or prestige for students and professional research staff, universities can collectively play a significant role in hastening the advance of open research and the broad benefits it delivers for the research community.