The latest Ithaka S+R report on agriculture scholars summarizes the diverse research being undertaken in across the agricultural sciences and suggests some ways forward for the agricultural information community. One key theme in the report focuses on the work undertaken by researchers in the agricultural sciences around data management. The breadth of the field is reflected in the variety of methods and data generated by scholars, ranging from genetics to economics to field-based biology. Across these domains, a common refrain emerged in the report suggesting that agriculture scholars struggle with data management issues. They are not sure what to do with their data upon collection, how to evaluate it for long-term preservation, and where to deposit it. While some reported using institutional repositories, others were unsure about the potential for funder mandates around data in the future and what these might mean for their future data management decisions.

As a member of the Ithaka S+R project team at the National Agricultural Library, I observed that many researchers are increasingly focused on issues around scientific data; this was reflected in our work as well as that of the other participating land grant university libraries. One of the reasons for the prevalence of the data management conversation could be the 2013 memo by then-director of the US Office of Science and Technology Policy, John Holdren, on increasing access to federal scientific data. This directive instructed federal agencies with annual research & development budgets of above $100 million to develop a plan for increasing public access to both data and publications resulting from federally-funded scientific research.

The 2013 Holdren memo was also the subject of a paper recently published by myself and three co-authors titled “An Analysis of Federal Policy on Public Access to Scientific Research Data.” In this paper, we read and analyzed 19 agency plans for increasing access to federal scientific data, comparing approaches along a number of themes. One of our primary findings was the widespread adoption of requirements for intramural and extramural scientists to submit data management plans (DMPs) along with research proposals, with 17 of 19 plans discussing DMPs in some detail. Federal scientific agencies appear to agree that data management plans are a key component of their broader strategy for providing better access to more datasets resulting from government-sponsored research. However, this does not square neatly with the recent Ithaka S+R report, which demonstrates that while agriculture scholars are aware of DMP mandates from funders, they are unclear about what they will include and how much (if at all) they will judged based on their follow through with actions specified in these plans.

As of this writing, the OSTP does not have a director and the future of the plans to increase access to government research data are unclear. What is certain is that, across scientific communities, data management is no longer solely within the purview of librarians and information professionals. For researchers who view their data as part of their overall scientific contribution, worthy of sustainable management and potential preservation, issues of research data are closely connected to other big questions for scientists such as the value of datasets for promotion and tenure, the role of data reuse in research and teaching, and additional requirements to obtaining grant funding for research. As Ithaka S+R continues its work documenting the research practices of scholars across the academy, it will add additional insights into the role of data management in different communities. Additionally, I hope that my latest paper advances research and conversation around the role of governments in scientific research. What policies and requirements are useful for scientific communities, and what might only work for a small subset of researchers? Should government funding agencies enforce what applicants outline in DMPs? As more funding bodies require these plans, we will get a better idea of what effect they have on deposits in repositories and on the ability for others to access and reuse data.