Date: 
25 February 2022

Data management planning has become a central concern for publicly funded research, with both funders and academic institutions acutely aware of the ethical and legal responsibilities that come with conducting rigorous and transparent research. The changing landscape about data protection has made this a tricky area to navigate.

On 14th and 15th February 2022, SSHOC facilitated a 2-day, interactive workshop on optimal handling, organising, documenting and enhancing research data, with a particular focus on how to write data management plans. Over 30 participants from across Europe came together to listen, discuss and ask questions about effective research data management. Presenters from SSHOC project partner UK Data Service (UKDS) gave information presentations, followed by interactive sessions where participants critically discussed case studies and personal experiences of data management in social sciences.

 

Image 1: Audience’s research interests

 

Drafting up Data Management Plans – Day 1

Cristina Magder, Data Collections Development Manager at UKDS, opened the workshop with a comprehensive look at the structure and standards of data management plans (DMPs). She introduced a practical checklist to help strategically address key issues within DMPs.

 

Image 1: Data Management Plan Checklist: Planning

 

Particular attention has been given to the FAIR (Findable, Accessible, Interoperable, Reusable) principles and their role in Data Management Planning. Some funders have now moved to clearly requesting evidence on how the FAIR principles will be addressed and implemented during the research data lifecycle. The Horizon Europe DMP template has been given as an example to illustrate the emphasis on FAIR and how good data management planning enables researchers to meet these requirements.

Cristina Magder then moved to the ethical and legal considerations of data management. As part of this discussion, she noted the duty of confidentiality and standards set by GDPR and national data protection legislation which can impact how consent is gathered and how personal data is stored and accessed. Participants asked questions around copyright restrictions and how to apply these standards in difficult research contexts, such as when using audio-visual data.

For the second half of the day, Anca Vlad, Research Data Services Officer, introduced how to prepare data for publishing and the principles of data curation. She focused on how much more effective data management is when you can plan and act as you go, rather than aiming to retrospectively clean the data. In preparing data, Anca introduced a 3-prong strategy to protect participants, which includes gaining thorough consent before collecting data, having a clear plan of anonymisation and treating the data as you go, and controlling who can access it. These three complementary strategies all form a comprehensive plan to protect identities and comply with expectations around data management.

 

 

Anca Vlad noted the importance of keeping the documentation and ensuring that data has enough context for an added-value of the particular data. She also signposted to open source templates that can be re-used, such as the UKDS Model Consent Form. The key components of good data management planning are quality assurance and security of data during the entire research lifecycle. The speaker presented strategies to ensure data is kept secure, which includes a digital back-up strategy that is shared securely and disposed of in a safe manner (e.g. collaborative environments).

Good data management should also include good costing for research projects. On average, to prepare and collate materials for deposit, two to three weeks should be costed into a typical two-year research grant application. Data and accompanying materials should be prepared throughout the research project as this has been proven to be more cost effective. The UK Data Service Data Management Costing Tool and Checklist is an open source resource created to help formulate research data management costs in advance of research starting.

 

Challenges to sharing data and success stories – Day 2

On day 2, Maureen Haaker, Senior Qualitative Data and Training Officer at UK Data Service, presented what archived datasets look like, how to assess these for suitability for your research, and how to find data. She noted the ethical responsibility to use existing data, when it is relevant to the aims of the research project, before embarking on creating new data. When searching for suitable data, she encouraged participants to use CESSDA’s Data Catalogue or check for data papers in data journals, like Research Data Journal for Humanities and Social Sciences.

The participants then discussed key issues that should be addressed within a DMP for three research scenarios (see the image below). The discussion revolved around consent for specific types of data, proprietary formats, sharing among research teams and concerns related to sharing.

 

 

The workshop and day 2 wrapped up with a final presentation from Cristina Magder on the challenges to sharing social science data. The participants were asked to discuss their concerns of data sharing, which, as noted by the speaker, were very similar to the challenges raised in the 2018 Springer Nature Whitepaper: Practical challenges for researchers in data sharing. According to data, most researchers were struggling with organising data in a presentable and useful way (46%), followed by copyright and licensing uncertainties (37%), not knowing which repository to use (33%), lack of time to deposit data (26%) and costs of sharing data (19%).

Source: Astell, Mathias; Admin, Springer Nature (2018): Infographic - Practical challenges for researchers in data sharing. figshare. Journal contribution.

Image 3: Barriers to sharing data, as recorded by workshop participants.

 

The SSH Training Discovery Toolkit includes a plethora of materials that researchers can use to ease their data sharing journey. Cristina Magder noted several strategies of identifying appropriate responsible repositories, the range of licenses available for deposited data, and the extent to which participants want their data shared. She demonstrated that the barriers to data sharing can often have solutions, especially when exploring options for publishing data. As argued in Kuula, A. (2011). Methodological and Ethical Dilemmas of Archiving Qualitative Data. IASSIST Quarterly, 34(3-4), 12, retrospective consent for sharing is a powerful tool in a researcher’s toolbox. Out of 169 research participants contacted by the Finnish Social Science Data, 165 have consented to data sharing (98%). For researchers conducting secondary analysis keeping track of data use by using a variable log and negotiating permission to republish are a must. Alternatively, well documented code and syntax can be freely shared and enable reproducibility and reconstruction of derived data.

 

Want to know more?

If you would like to have a look at the information presented in this workshop, we invite you to take the time and watch the workshop recordings (Day 1, Day 2) and check the presentations (Day 1, Day 2).