Academic research is usually based on research data, which is formed, processed and analyzed. Before the Internet, the storage, preservation, and recycling of research data depended on disciplinary practices, research groups, and individual researchers, but the situation has changed since then. In 2013, Kone Foundation began to recommend that its grantees archive their research data for further use. Since then, research data have become almost comparable to publications. Research data repositories and dedicated journals have been established.
It is necessary to distinguish between different types of research data. Raw data is data that is constantly updated during the data collection, or it can be produced by software. Working data can already be referenced, and for that to be the case, it must be organised into versions. The most structured form of data is data output, which in principle does not change any more, but can of course become part of a broader research material. (These distinctions are based on a presentation by Heidi Laine from the CSC at an event of the Finnish Association of Foundations in May 2021.)
Today, researchers are instructed to manage their data well. This should make the data more valuable by rendering it interoperable, easier to find, and harder to lose, more understandable, and reusable. The abbreviation FAIR (findable, accessible, interoperable, reusable) is used in English. The Council of the European Union and the European Commission encourage good data management in all research. In February 2020, the European Federation of Academies of Sciences and Humanities published a special recommendation on research data management in the humanities.
Data management plans have found their way into funding applications, too. For instance, the Academy of Finland requires such a plan, because “[r]esearch data and research publications are among the most important outputs of publicly funded research”. One can note that research data are mentioned before publications. In Finland, the DMPTuuli web service is supposed to support academics in preparing data management plans.
Now, could data management plans help avoid situations like the one I encountered with one of my sets of research data? The printed book version of my dissertation was published in the series of the Finnish Society of Sciences and Letters in 2004. I decided not to publish my research materials – the photographs – in the book, but on the university website, because I considered it more useful for the academic community. I only shared the book online a few years later. The photographs, however, disappeared from the Internet in the late 2010s, possibly due to a website renovation. Fortunately, the situation was remedied by a project in which the same material was photographed again and archived in accordance with the principles of open science on the website of the Italian National Research Council.
At least for the time being, Kone Foundation does not require research grant seekers to submit a data management plan. First and foremost, our evaluators decide how data management affects the success of a funding application. We continue to recommend the openness of research materials to research grantees, but we do not provide further guidance. However, it is also in the interest of the funder that research materials are not wasted. In my view, researchers need to learn to pay attention to data management, and the instruction is primarily a task of graduate education. The fact that more and more research is actually being done in groups can also facilitate this development in the humanities, if a member of the group can take on a bigger role in data management, and also gain merit from it.
When research funders define the elements that must be included in funding applications, we must ask – ourselves and the rest of the academic community – which elements are necessary? How can we maintain a balance that also leaves room for the main contents of the study? This is the big challenge that a research funder has to constantly consider, because no one wants to make application and evaluation processes more laborious than they already are. If, on the other hand, it starts to look like researchers are not gaining enough understanding of data management in their research organizations, then we need to think about how the situation could be improved together.