guest post by Zsuzsa Szabó, University of Debrecen, OpenAIRE NOAD Hungary Research data management is an emerging issue worldwide. There are several initiatives within the European Union and its member states to implement data management plans and policies. Open access to publicly funded research results published in scholarly journals is an underlying principle in Horizon 2020.The third phase of the OpenAIRE project series, OpenAIRE2020 focuses on research data management, data archiving and sharing practices. The Open Research Data Pilot aims at improving and maximizing access to and re-use of research data generated by selected Horizon 2020 projects.
Research data has become an increasingly hot topic not only in the international but also in the institutional agendas. Many research institutions and universities have adopted open access policies and recommendations on dealing with research data of Horizon2020 projects. The rising demand from international funders and EU regulations regarding this issue also has a high impact on Hungarian research institutions. The University and National Library University of Debrecen is the National Open Access Desk (NOAD) for Hungary in OpenAIRE2020. As a first step in mapping research data management practices in the country, the present study covers research data management habits, data types, archiving practices, data accessibility and re-usability, relevant legal and ethical questions at the University of Debrecen.
Our survey was conducted using the software Evasys and was sent out in March 2017 via the university mailing system three times within a three-week-time period. We targeted all of our 14 university faculties. We received 237 responses out of 2653 academic staff. Almost 10 % of the researchers answered. One third of the respondents were PhD students.
There were five main question categories:
data types and formats
data storage, archive and data loss
ethical and legal issues
data accessibility , share and re-use
We also asked researchers about their own views on institutional help and if they have any suggestions regarding data management. Survey results were summarized on a poster at the COAR Annual Meeting in Venice 2017. Major findings: Respondents The majority of completed questionnaires came from life sciences (35,86%) and medical sciences (23,68%) followed by social sciences (15,19%), humanities (12,24%) and agriculture (7,17%). The remaining less than 10% of the respondents were represented by technical sciences and arts. Research data types and formats Similar to international trends, almost all of the surveyed researchers mostly generate text (95,38%) and graphics (68,35%) during their research projects. One third of the researchers also create databases, software applications and structured text files. Regarding the generated data size per year showed that 44% of all data from different disciplines do not exceed 50 gigabytes of storage capacity. There was only one respondent (representing natural sciences) who stated a need for a massive ( above 1 Pb) storage capacity. Big storage ( between 1Tb and 1 Pb) was required from the medical sciences (44,4%) and natural sciences (55,6%). Data storage, archive and data loss Most of the researchers use multiple data storage options, most commonly their own laptops (85,65%) external hard drives (64,98%) or institutional computers (54,01%). Only one third of the researchers store their data in free cloud services and very little, less than 4% use institutional or external data repositories. When asked about data archiving practices, 95,2% of the scientists think they are responsible for archiving their data - more than half of them doing it for indefinite time period. 52 % of the researchers have never experienced data loss. Although the majority (87%) attach either standard or own documentation with the data, over 57 % of them do not use any descriptive metadata standards. Ethical and legal issues Regarding the ownership of the data 44% of the researchers believe that the owner is their own institution while 43% believes the data belong to the researchers. It is interesting to note that 30% of the respondent do not know who owns their research data. The result was similar with the question “what happens with the data when the researcher leaves the institute:” one third believe it stays with the institute, 25% take data with themselves and over 32 % couldn’t answer this question. This issue calls our attention to the lack of clear legal and ethical guidance within the institution. Data accessibility, share and re-use Despite the fact that researchers do not like sharing their data, the survey produced a positive attitude in our institution. 47% of the research data can completely and 37% can partially be re-used according to the scientist. The researchers are willing to share their data either with the research community (54%) or with their own institution’s researchers (35%). 12% of the respondents stated their data are openly available. Regarding the multiple answer question “how to share their data” more than half (58%) would choose to share via email, 34% would use a cloud service, 30% would share through an external hard drive, 24% via their institutional repository, 16% through the institutional or personal website and only 8% via intranet. Lack of ethical and legal documents within the institution reflect in the negative answers rather researchers are following any policy regarding data sharing. In the last segment of our survey we asked researchers what they expect from our university regarding research data management. 41% would like to have a data management course integrated in BSc and MSc curriculum, 38 % would expect a development of a single institutional policy and 38% would like to have an institutional repository for this purpose. Despite some negative tone most of the researchers would highly appreciate a stable, trustworthy institutional repository with associated data management support and IT help. Our respondents also showed a high demand and interest in classes and training on subject specific data management and they also showed interest in a cooperation of the development in a customized data management plan. Update: the original file is available on Zenodo: https://zenodo.org/record/800523#.WXCKP4iGP4Z