Always on My Mind: Sustainability and Data

by Patricia Hswe, Andrew W. Mellon Foundation

Since it’s “Love Your Data Week,” I feel it’s appropriate to kick off this blog post with lyrics from a love song made popular by Willie Nelson, “Always on My Mind.” Most of its first stanza – but, admittedly, not much more – could easily be about a researcher and her data:

Maybe I didn’t love you
Quite as often as I could have
Maybe I didn’t treat you
Quite as good as I should have
If I made you feel, oh, second best . . .[1]

A key motivation behind the outreach and engagement encompassed by “Love Your Data Week,” however, is to head off regrets and misgivings about data, which for many researchers constitute the heart of their work. Loving data means learning to be mindful and responsible about the integrity of your research, not to mention, as well, your future researcher self. Even a few days from now, do you really want to be playing “detective” with your data, ferreting them out from your hard drive, or from the cloud? Or trying to parse them, because you put off labeling and describing them in meaningful yet standardized ways? Or, worse yet, trying to salvage them because of server failure or a security breach?

It’s well known that many federal funding agencies require a data management plan (DMP) from their grantees, a mandate intended to decrease, if not eliminate, scenarios like the ones posed in passing above. In my previous position as an academic librarian managing user services for the campus’s institutional repository, I had a leadership role in providing research data management guidance and support. It would not be an exaggeration to say that on some days, data was always on my mind. Now, as a program officer for scholarly communications at the Mellon Foundation, I work on the funder side of things – although I am no less concerned about data longevity and use.  One factor to note is that we make grants for scholarly communications projects that are grounded in the humanities; these efforts typically do not produce data on the scale that projects funded by other organizations, such as the National Science Foundation, routinely generate and strive to support. Nonetheless, a cursory search in our grants database shows that, over the course of several years, scholarly communications has made 40 grants, totaling $19.21M, that address or involve some aspect of data (e.g., data privacy, data curation, a data dashboard) or different types of data pertinent to humanities fields (e.g., archaeological data, linked open data, annotations).

While we do not require grantees to submit a DMP, we do ask them in our guidelines (PDF) to address the issue of sustainability. Grantees are asked to give, “[a]s appropriate, an account of how the organization will ensure the longer-term sustainability of project results and/or institutional changes supported by Foundation funding.” Our guidance for the intellectual property section of the proposal also addresses sustainability, asking “how the organization will ensure the long-term sustainability of the digital and/or software products.“ The language in our sustainability prompt is non-prescriptive because of the wide range of project possibilities, and because much depends on the context of the proposal and the aims and deliverables of the project. At the same time, a grantee might view the statement as an invitation to be more intentional about the sustainability of a project’s accomplishments from the very start, taking into account – as relevant – any data, workflows, software, and other data-intensive or data-dependent outputs that are produced. Where possible, we encourage open sharing of research outputs. Active usage of project deliverables, such as data and technology, especially if well documented, can create a demand that, in turn, helps sustain them. As with strong DMPs, a strong sustainability section evidences that the grantee has thought through the future prospects of the work, perhaps confirming continuing engagement, support, or collaboration of some kind, or the resolve to seek such as part of the industry of the project. Of course, some projects, being small or restricted in scope, may not need to continue past a certain point, making sustainability considerations less burdensome; an assertion of such is not unlike an NSF proposal in which a DMP is deemed unnecessary by the grantee, who must state so and why.

If the sustainability section gives grantees pause, then I view the reflection as favorable – and the prompt a success up to that point. The last thing I would wish upon grantees is the regret that infuses “Always on My Mind”: “Little things I should have said and done / I just never took the time.” In other words, say and do those little – even big things – and please take the time!  Your data users and re-users will thank you.

[1] Willie Nelson. “Always on my mind.” Always on My Mind, Columbia, 1982, Spotify,

RWJF loves your data

We love your data, and we want to help encourage its use developing programs and policies that improve population health, well-being, and equity.

by Oktawia Wójcik, PhD, and Margaret Tait, Robert Wood Johnson Foundation

At the Robert Wood Johnson Foundation, we’re working to build the knowledge base of what works for health.  As the largest philanthropic organization in the U.S. focused solely on health, we think a lot about data and how it can be used to identify gaps, measure progress, and inspire action. We’re interested in using data, big and small, to tackle issues of health equity, health system integration, and community transformation. And when we think about data that is related to health, we are thinking about more than just information collected in medical records, we are thinking about social determinants of health data like education, transportation, and criminal justice.

We’ve spent the last two years learning ways to increase openness and transparency in research, with the goal of making the research that results from our investments, and the investments of others, more accessible to those who need it.  Doing so ensures that a wide net of health services researchers are aware of the economic analyses of population-based payment methods and that city planners have the research showing life expectancy variation by location.

Openness is just one part of the equation, though; and funders have to be intentional to ensure the research we support is of good quality, rigorous and representative.  To that end, we see the role that funders can play in data stewardship as one that is twofold: to encourage and challenge. As we develop programs and consider key research studies, we want to encourage the use of data, both quantitative and qualitative, to inform all aspects of an approach and we want researchers to use data at various levels, individual and population. We recognize that there are challenges associated in working with individual level data, such as privacy and confidentiality that need special consideration. As funders, we are keenly aware of and intentional in supporting data of high integrity and quality, as well as methods for the analysis of this data that are shared with others as a standard practice.

In collaboration with the Centers for Disease Control and Prevention (CDC) and CDC Foundation, we support the 500 Cities Project.  This project uses small area estimation methods to provide data on 30 health indicators at the census tract level in 497 of the largest U.S. cities, as well as a city each in West Virginia, Vermont, and Wyoming, covering all 50 states.  This is the first time this amount of data had been available at a city and census tract level, and we hope it is just the beginning. Our vision is that this data will help researchers, practitioners, and policy makers identify issues and encourage collaboration on solutions that will promote more opportunities for better health.

The Foundation has also long supported the Interuniversity Consortium for Political and Social Research (ICPSR) and requires that publicly available datasets that result from our grant making are deposited here in the Health and Medical Care Archive.  ISCPSR has been an important part of RWJF’s work for decades, but recently we’ve been thinking about additional steps we could take to promote openness in the research process.  And, we’d be remiss if we didn’t take the opportunity to ask fellow data lovers: what role do you see for Foundations in data stewardship?