Statements on research data management in funding proposals
©
StartupStockPhotos | Pixabay
Many funders require your funding proposal to contain statements about the handling of research data. There may be a separate chapter dedicated to this matter, or you are obliged to submit a data management plan (DMP). Specifications regarding form and content vary between funders, disciplines and even between funding programmes and tenders.
Some statements reflect our personal assessments based on our counselling experience, or take into account informal communication with representatives of funders and project management agencies. We progressively strive to add further details to the information and keep it up to date. For this purpose, continued exchange with funders and colleagues from other research organisations is key.
What you should consider in general
-
General tips for proposal chapters on research data management (RDM) and for data management plans (DMP)
-
Take the RDM chapter or the DMP seriously and do not postpone the draft until submission deadline is approaching! The importance of the RDM topic is increasing constantly and hence, its impact on the reviewers` verdict is growing as well.
-
Stay on topic. It is ONLY about the handling of DATA. Information on your planned journal articles and attended conferences or your opinion on the significance of open science in general is rather out of place here.
-
Try to be as concrete as possible and avoid general phrases. Example: Instead of “We regularly backup our data” a statement like “The data is backed up daily on servers of the IT department of LUH via the service Backup & Restore” is preferable.
-
Use short sentences (or just key points) carrying a clear message.
-
Be realistic. Only promise what you can actually put into practice. If you mention concrete services, infrastructure and so on, check beforehand whether these in fact meet your requirements.
-
Carefully read the proposal guidelines, the DMP template or the funding tender to find out what kind of information the funder expects exactly. Check if your text actually covers every point mentioned there!
-
-
Data management requirements of major funders
Practically all funders expecting RDM statements demand that you comply with the so-called FAIR principles. Thus, your data should be findable, accessible, interoperable (with other data and technical systems) and re-usable for third parties.
"Do I have to publish all my data?"No, no funding body expects you to always publish everything, nor that you always keep everything. However, should always decide consciously and well-founded which data to keep or even publish and which to delete or restrict. Ethical and legal requirements may be good reasons to forego publication, for example. The guidelines 13 and 17 of the DFG codex on safeguarding good research practice take this into account. In EU projects, the principle "as open as possible, as closed as necessary" applies, which boils down to basically the same idea.
In your proposal or in an accompanying data management plan you set out, how exactly you are going to ensure compliance with the FAIR principles. The background for this requirement is that data generated with public funding should be processed and analysed as thoroughly as possible, even beyond the original project context. Furthermore, research results based on FAIR data are easier to understand and verify.
Important!Reviewers are usually encouraged to take into account discipline-specific standards when evaluating proposals. You should hence inform yourself about “best practice” in your field and find out whether discipline-specific RDM guidelines exists. You can find such information in online portals of disciplinary organisations and NFDI consortia, and on the RDM website of DFG.
TIP: The DFG portal “Research Integrity“ includes a feature to show discipline-specific comments on each of the “Guidelines of Good Research Practice”.
-
RDM support services at LUH
At LUH, a wide range of support services for data and information management is available. The Guidelines for Handling Research Data at Leibniz Universität Hannover offer basic recommendations for action. The website on research data management provides detailed information, materials and tools. The Team Research Data Management offers project-accompanying advice and organises introductory and in-depth courses on research data management, which can also be adapted to specific disciplines. In addition, exchange on RDM topics takes place in regular networking events.
The IT centre at LUH (LUIS) provides infrastructure services, central hosting of RDM-relevant software and technical support. It also hosts an institutional research data repository, in which LUH members may publish their research data and generate a corresponding DOI.
TIB offers discipline-specific research data repositories and RDM consulting services in the scope of the disciplinary information services "Mobility and Transport" and "BAUdigital". The DOI service functions as a point of contact for institutions that wish to generate DOIs for research data archived and published in their (institutional) data centres. In addition, research data related to a doctoral thesis may be published in the LUH Text Repository operated by the TIB, together with the thesis itself.
German Research Foundation (DFG)
-
What kind of information on RDM do they demand and in which form?
The DFG requires an RDM proposal chapter in almost all of its funding programmes. You can find details of what should be included in this section in the relevant application guidelines or templates. They usually refer to this checklist, which is divided into the following six sections:
- Data description
- Documentation and data quality
- Storage and technical archiving during the project
- Legal obligations and conditions
- Data exchange and long-term data accessibility
- Responsibilities and resources
The following points are also highlighted (the text is taken from the guidelines for Individual Research Grants):“Please also describe how the institutions involved in the project will contribute to data and information management.
If you have already provided more detailed information on the handling of research data in an explanation as part of your preliminary work, work programme or elsewhere, you may refer to those descriptions and limit yourself to supplementary information at this point.
Should your project not use or generate data to a relevant extent, please explicitly state this to be the case.”
You can also apply for funds for RDM-specific expenses exceeding the usual basics that every university is expected to provide. Such expenses could, for example, consist in costs for additional personnel tasked with elaborating data for optimal re-use, or costs for data publishing. If you plan to commission external service providers, check whether they have a schedule of fees or get an individual quote in time.
Please take into account the information and especially the FAQ at the DFG website on research data.
-
How extensive should the RDM chapter be?
That depends on funding programme, project size and discipline or research topic. DFG does not specify on this point. In case of individual research grants and sketches (RTG, CRC), we recommend dedicating between half a page and a whole page to the chapter. As far as full proposals for larger projects are concerned, the chapter may be significantly longer, because you should be more specific on issues like exchange of data between partners (access rights/infrastructure), uniform standards (documentation/file formats etc.) and concepts for privacy protection, if applicable.
-
How should the RDM chapter be structured?
Again, there are no fix specifications. We recommend the following structure:
- Announce that you will elaborate a data management plan at the start of the project (optional, but highly recommended).
- Information according to the sections of the DFG checklist:
- Data description
- Documentation and data quality
- Storage and technical archiving during the project
- Legal obligations and conditions
- Data exchange and long-term data accessibility
- Responsibilities and resources
Depending on the project, it may be advisable to change the order of the information or to omit points that are not relevant. Do not forget to apply for funds to cover cost for RDM-related measures beyond the university's basic services and infrastructure!
-
Do they check the implementation of measures described in the proposal?
There is no routine audit of every project. However, when you apply for an additional funding period, reviewers may have a closer look at your original announcements and check whether you actually put these into practice. An audit may also take place in special cases, e.g., when scientific misbehaviour is suspected.
European Union (EU)
The exact requirements may vary between programme lines ("actions"). The following statements refer, in first place, to the Research and Innovation Action (RAI) and the Innovation Action (IA) since by far the most funding applications are filed in the scope of these two programme lines.
-
What kind of information on RDM do they demand and in which form?
In Section B (Project proposal – Technical description) of the Standard Application Form (HE RIA, IA), under point 1.2 (Methodology, page 8), you should address the topic of Open Science, which also includes the publication of research data. In particular, you should specify the types of data you are working with, how you are taking the FAIR principles into account, and what resources (staff and funding) you are budgeting for the processing and long-term preservation of the data.
If your application is successful, you must also submit a data management plan (DMP) within six months after the project started. This document should be updated regularly and supplemented where necessary throughout the project, but at least at the mid-point and at the end of the project period. The data underpinning research results must be published open access ‘as early as possible’, i.e. normally before the end of the project period.
You can find further information in the annotated standard contract, known as the ‘Annotated Grant Model Agreement’, in Annex 5 (HE Communication, Dissemination, Open Science and Visibility) on pages 382–383, and in the explanatory notes under point 7, ‘Open science: Research data management’, on pages 389–392. Furthermore, Chapter 17 “Open Science” of the Horizon Europe Programme Guide sets out the requirements for research data management and data management plans (pp. 45–53).
-
How extensive should the RDM statements in the application form and the DMP be?
The information given in the application form may comprise up to one page. There is no such limit for the data management plan generated during the project, however. Its extent conforms to the requirements of your project. In case of interdisciplinary and multi-institutional joint projects you should calculate with no less than three pages, since you need to consider diverging local circumstances and disciplinary data management requirements. However, about ten pages are common.
TIP: Unless all RDM is centralised with one project partner, keep the DMP for the overall project relatively brief and general, and set out the details in partner- or sub-project-specific sub-DMP. This approach helps to maintain clarity and enables a better description of local conditions (e.g. locally available IT infrastructure).
-
How should the RDM statements in the application form and the DMP be structured?
The Programme Guide (Chapter 17, ‘Open Science’) sets out which measures and activities form part of the data management and should be described (pages 47–52). It is advisable to follow its order, i.e. to begin with a description of the data types, move on to explain how it will be ensured that data is findable, accessible, interoperable and reusable (FAIR principles), and finally address the resources allocated to data management and archiving. However, this order is not mandatory. It is important, though, that you cover all the mentioned points in substance, insofar as they are relevant to your project - briefly in the application, and in detail in the DMP.
There is an MS Word template available for creating data management plans in Horizon Europe projects. It provides a chapter outline and guiding questions to help structure the content. However, you will notice some points where certain information relates to more than one chapter. In such cases, you do not need to repeat yourself. Instead, you should refer to the relevant sections where you already covered the topic (“see…”).
Hint for ERC-GranteesFor individuals receiving excellence funding from the European Research Council (ERC), the official guide “Open Research Data and Data Management Plans” is relevant. It details the expected contents of a DMP written in the scope of a respective project and lists recommended tools and repositories. There is also a special DMP template for this funding programme.
-
Do they check the implementation of measures described in the proposal and the DMP?
Your statements in the grant agreement form part of the funding contract and are therefore legally binding. A DMP is a deliverable, and a first version has to be submitted within the first six month of the funding period. Keep it up to date and include the current version in your periodical reports and in the final report.
Completeness and quality of the RDM chapter and the DMP will be considered during the evaluation of midterm reports and final reports. To our knowledge, there are even special reviewers who check whether a DMP is adequately concrete and complete and then report their verdict back to the respective project officer. In case of critical assessments, it is up to this officer if and how he or she requests improvements.
We currently do not believe that the practical implementation of the measures described in the plan is regularly verified on site. However, you should expect a closer examination on samples of everything that can be checked online (e.g. data publication in compliance with the FAIR principles). An extraordinary audit may also take place in case of complaints.
Federal Ministry of Education and Research (BMFTR)
-
What kind of information on RDM do they demand and in which form?
BMBF passed an “Aktionsplan Forschungsdaten” (action plan research data), which outlines general aims of the political agenda. When it comes to funding of scientific projects, however, there are (still) no uniform requirements regarding data management. Instead, it is defined separately for each funding tender, if and to what extend RDM statements are mandatory. Requirements depend especially on the discipline(s) addressed by the tender. In case of disciplines with an advanced RDM culture, e.g. educational sciences, you generally have to submit with the proposal a data management plan (DMP) following discipline-specific standards, and additionally a “statement on generation of new data”.
-
How extensive should the RDM chapter be?
Again, this is defined in the respective tender. Take the “guideline on funding junior research teams in empirical educational science”, for example. It states in section 7.2.2, point D (attachments) that the “statement on generation of new data” must not exceed 1.500 characters (spaces included). The DMP may be up to 6.000 characters long, which roughly translates into 1.5 to 2 pages.
-
How should the proposal´s DMP be structured?
Carefully read the tender and check whether it specifies on DMP structure. In case of educational sciences, for instance, current tenders generally refer to the “standardisierter Datenmanagementplan für die Bildungsforschung” (Stamp). If there are no specifications, you can follow our recommendations on how to structure RDM chapters in DFG proposals.
-
Do they check the implementation of measures described in the proposal and the DMP?
We do not have information on this point, currently.
Volkswagen Foundation
-
What kind of information on RDM do they demand and in which form?
Applications for Volkswagen Foundation programs require RDM information in the form of a data management plan. As an assist, the foundation provides a template (basic data management plan). It contains seven questions on data origin, data types, data organization and data publication. If it is already determined in which repository the data will be published, the requirements of the repository (for example, regarding the format of files and metadata) should be considered. In addition, it makes sense to at least outline the intended data management in the proposal itself, even if this is not explicitly required in the call.
Further recommendations and principles of the Volkswagen Foundation on Open Access, Open Data and Open Source are set out in the Open Science Policy. These should also be taken into account.
TippYou can apply for additional funding for the preparation and publication of outstanding data generated within the scope of a funded project via the programme “Data Reuse - Additional Funds for the Preparation of Research Data”. Applicants have to describe the planned cooperation with a suitable disciplinary repository in which the data will be published.
-
How extensive should the data management plan be?
The Volkswagen Foundation does not specify a general upper or lower limit for the extend of the data management plan. However, there could be corresponding specifications in the respective calls. The expected level of detail of the given information may vary according to the type of data and the research discipline. Existing community-specific standards should also be taken into account. As a rough orientation, we recommend a length of about two to five pages, unless specified otherwise in the call.
-
How should the proposal´s DMP be structured?
In principle, Volkswagen Foundation recommends the use of the basic data management plan, for which it provides a template.
-
Do they check the implementation of measures described in the DMP?
No RDM-related audits take place during the project. However, Volkswagen Foundation beneficiaries are obliged to submit a report after completion of their project, and this report is expected to contain information on RDM. The information given there should at least not grossly contradict the intentions originally described in the DMP.