Research Data Management in India A Pilot

Research Data Management in India : A Pilot Study By Dr Nishtha Anilkumar Head, LIS Physical Reseach Laboratory 7 June 2017 LISA 8 1

Outline of the presentation ü ü ü ü 7 June 2017 Introduction Need for RDM Stakeholders Challenges Data Policy Pilot Survey Data Centres Conclusion LISA 8 2

Physical Research Laboratory Navrangpura, Ahmedabad 380 009 7 June 2017 LISA 8 URL : http: //www. prl. res. in 3

The Physical Research Laboratory was established to fulfill the vision of Dr. Vikram A. Sarabhai: “Countries have to provide facilities for its nationals to do front rank research within the resources that are available. It is equally necessary having produced the men who can do research, to organise task oriented projects for the nation’s practical problems. ” The first step to set up PRL was taken in 1947 when Dr. Sarabhai established a laboratory for Cosmic Ray Studies in his home “RETREAT” 7 June 2017 LISA 8 4

7 June 2017 LISA 8 5

Library Collection • 21, 000 books • 35, 000 bound volumes, • 1500 videos/CDs/DVDs • 174 journals out of which 160 are online • Reports, Maps, Reprints 7 June 2017 LISA 8 6

Library Infrastructure LMS server (Lib. Sys) RFID security system Institutional Repository server (GSDL) 10 PCs for the staff 12 PCs for OPAC viewing 3 Photocopying machines 4 printers 3 Barcode Printers 7 June 2017 LISA 8 7

Library Services • • 7 June 2017 Circulation OPAC (Reservation) Library Homepage (Electronic Journals Access) Interlibrary Loan & Document Delivery Service Assisting in procuring books for book grant Plagiarism check SDI service Photocopying Service LISA 8 8

Research Data Defined as “the recorded factual material commonly accepted in the scientific community as necessary to validate research findings” (OMB Circular 110) Research data covers broad range of information like documents, spreadsheets, field notebooks, diaries, audio tapes, video tapes, images, spectra, models, algorithms, scripts, protocols, workflows, software, standard operating procedures, methods It does not include preliminary analysis, drafts of scientific papers, plans of future research, peer reviews, communication with colleagues, trade secrets, commercial information, personnel and medical information. 7 June 2017 LISA 8 9

Research Data Management • Sometimes known as Research Data Curation • Consists of two components - data preservation and - data access • Collecting and organising the data which is part of the research outcome in such a manner so as to facilitate easy access and re-use. 7 June 2017 LISA 8 10

Need for Data Management • Most data are produced or gathered as part of publicly funded research, so it needs to be transparent, accountable and available • In many countries now, major funding agencies mandate that applicants submit a data management plan (DMP) as part of their research proposal • Data life cycle provides an overview of the stages involved in successful management and preservation of data for use and reuse 7 June 2017 LISA 8 11

Data Cycle Source : Siyavula. com 7 June 2017 LISA 8 12

Stakeholders in RDM • Researchers • producers of data • hence 1 st stakeholder • reluctant to share Any endeavour to share that data will depend on the trust they have on the RDM unit 7 June 2017 LISA 8 13

Stakeholders in RDM • IT Services Currently, most researchers keep their data on personal storage devices without documentation, version control or back up Robust IT infrastructure supports advanced data acquisition, storage, management, security, integration, mining and visualization as well as other information processing services 7 June 2017 LISA 8 14

Stakeholders in RDM Library Services Library is very well positioned to carry out RDM v Adheres to standards-based information organization v Information management skills that librarians have, such as assigning metadata to the information item for easy retrieval 7 June 2017 LISA 8 15

Library as a stakeholder v Libraries have earned a trusting relationship with researchers v General perception of library is “safe, sustained and trusted unit” for long term document/data preservation v Most libraries have experience with copyright issues v. Library can help the researchers in depositing their data to international subject repositories 7 June 2017 LISA 8 16

Challenges in RDM v Due to powerful computing technologies used, more and more number of researchers generate and use large datasets as part of the research process v Storing this data in a form that can be easily accessed, processed analysed is a very challenging activity for any research/academic institute v Datasets are potentially fragile, being vulnerable to storage failures and technological obsolescence v Cost of developing infrastructure 7 June 2017 LISA 8 17

Competencies required Cox (2014) has suggested following competencies for the library personnel to carry out RDM effectively : - strategic understanding and influencing skills - knowledge of RDM principles and policy - understanding of RDM best practices - knowledge of institutional resources - knowledge of researchers’ needs - metadata skills - knowledge of copyright and licensing issues - knowledge of relevant technologies and processes - supporting the researchers in preparing the DMPs - creating a web portal for researchers to submit their data Library staff responsible for data archiving need to develop all the above skills for effective RDM 7 June 2017 LISA 8 18

More challenges for Library v Capacity and workload on existing shrinking staff v Persuading the scientists to deposit their data for preservation and sharing and v Convincing the management that library’s information handling skills are relevant for data management 7 June 2017 LISA 8 19

Data Policy v Many universities which have no funder mandates for DMPs still carry out data curation and have a data policy in place simply because it is a good practice v Having a data policy helps in clarifying many issues and acts as a guide for the staff carrying out RDM v Before framing a data policy, an institute needs to constitute a committee comprising of various stakeholders and chaired by director/dean 7 June 2017 LISA 8 20

Data policy elements Briney, et al (2015) emphasized that following points need to be explicitly mentioned while framing a data policy of an institute : 1. Why data repository is important ? 2. Who owns the data ? 3. Who is responsible preserving and organising the research data ? 4. In case of collaborative research, who will preserve the data ? 5. What data should be retained ? who decides which data to keep ? 7 June 2017 LISA 8 21

Data policy elements… 2 6. How open should the data be ? 7. How long the data is to be preserved ? 8. What steps will be taken so that hardware obsolescence does not lead to data loss ? 9. Which kind of metadata is required for different types of data sets ? 10. Which software will be used for retrieving the stored data sets ? 7 June 2017 LISA 8 22

Data policy elements…(3) 11. What are the ethical issues ? 12. How is the data accessed ? 13. How will the costs be managed ? if the funding is project based, then how will long term preservation be supported ? 14. What happens when the primary researcher leaves the institution ? 15. How will the data be acknowledged and cited ? 7 June 2017 LISA 8 23

RDM in India To know the level of awareness or involvement of libraries in RDM, a pilot survey was done Sample consisted of 25 research institutes representative of various Consortia – FORSA, DAE, DOS, and DST. Five academic institutes from IITs, IIMs, IISERs and one private university - were also sent the questionnaire Out of these 30 institutes surveyed for the pilot study, 15 responded with the filled-in questionnaire 7 June 2017 LISA 8 24

Respondents of the pilot survey 1. Bose Institute, Kolkatta 2. Indian Institute of Astrophysics, Bangalore 3. Indian Institute of Management, Ahmedabad 4. Indian Institute of Science Education & Research, Pune 5. Indian Institute of Technology, Gandhinagar 6. Information and Library Network, Gandhinagar 7. Institute of Plasma Research, Gandhinagar 8. Inter-Univ. Centre for Astronomy and Astrophysics, Pune 9. National Institute of Oceanography, Goa 10. National Remote Sensing Centre, Hyderabad 11. Nirma University, Ahmedabad 12. Raman Research Institute, Banagalore 13. Saha Institute of Nuclear Physics, Kolkatta 14. Space Application Centre, Ahmedabad 15. Tata institute of Fundamental Research, Mumbai 7 June 2017 LISA 8 25

Findings of the pilot study v Out of the 15 responses received, at nine institutes data archiving was done by the researchers themselves and not by library or other division v Out of these 9 institutes, two libraries plan to start this service for researchers very soon (IIMA & IPR) v In four institutes (INFLIBNET, IIAP, NIO and SAC) data is archived by other division Out of these 4, only NIO has a data policy in place v In two institutes data is archived by the library but data policy formulation is in process (Bose Institute & RRI) 7 June 2017 LISA 8 26

Result of the pilot study Findings show that in India RDM carried out by libraries is still at very early stage of development and will take a few years more to become an active area of work However, there are quite a few data centres in India in different subject disciplines to cater to the researchers, the funding agencies and the general public 7 June 2017 LISA 8 27

A few data centres ICSSR Data Service is culmination of signing of Mo. U between Indian Council of Social Science Research (ICSSR) and Ministry of Statistics and Programme Implementation (Mo. SPI) Set-up with an aim to support researchers, teachers and policymakers who heavily rely on high-quality social and economic data for their research World Data Centre for Geomagnetism, Mumbai Responsible for the compilation of final hourly absolute values from nine of the Indian magnetic observatories and deposition of this data to the World Data Centre The centre also hosts a database driven website to make datasets available online to the global scientific community. 7 June 2017 LISA 8 28

Data Centres… 2 Indian Space Science Data Center (ISSDC) Is the primary data centre for the payload data archives of Indian Space Science Missions. This data center, located at the IDSN campus in Bangalore, is responsible for the ingest, archive and dissemination of the payload data and related ancillary data for Space Science missions like Chandrayaan, Astrosat, etc. ICRISAT Dataverse Network ICRISAT performs crop improvement research, using conventional as well as methods derived from biotechnology, on the crops like Chickpea, Groundnut, Pearl millet, Sorghum and Small millets. ICRISAT's data repository collects, preserves and facilitates access to the datasets produced by ICRISAT researchers to all users who are interested in it 7 June 2017 LISA 8 29

Kodaikanal Solar Observatory • Solar observations at this observatory over the last 100+ years provide one of the longest continuous series of solar data • Simultaneous observations in different wavelengths make this data a unique one and suitable for multi-wavelength studies • Historical data which were on photographic plates has been digitized. The digitised data are available for use by the scientific community. 7 June 2017 LISA 8 30

National Informatics Centre NIC, under the Department of Information Technology of the Govt. of India, is a premier S & T organization, for promotion and implementation of ICT solutions in the government With the increased expectations from citizens for online services, the Data Centre requirements are growing NIC has set up state-of-the-art National Data Centres at NIC Hqrs, Delhi, Pune and Hyderabad and 30 small data centres at various state capitals to provide services to the Government at all levels 7 June 2017 LISA 8 31

Indian Oceanographic Data Centre (IODC) • IODC was established in 1964 • IODC plays a dual role - dissemination of data / information to the user communities and - assisting the data personnel in processing, validating, reformatting different types of data generated from the Indian ocean region 7 June 2017 LISA 8 32

Conclusion The pilot study survey shows that in India RDM carried out by libraries is still at nascent stage of development and will take a few years more to become an integral part of RDM activity in research and academic institutes. However, the scene of data centre set up in different subject fields looks to be very promising PRL library has given the proposal of setting up the infrastructure for data preservation. We are in the process of framing a data policy and are engaging with the researchers to discuss about the benefits of data archiving for data access. We have a long way to go, but first step is taken… 7 June 2017 LISA 8 33

Institute Section reponsible Data policy BOSE INSTITUTE yes library being planned INFLIBNET yes other being planned IIAP yes other no policy IIMA no library (being planned) no policy IISER, Pune no no no policy IITGN no no no policy IPR no library (being planned) no policy IUCAA no no no policy NIO yes other yes NRSC no no no policy RRI yes library no policy SAC yes other no policy SINP no no no policy TIFR no no no policy Nirma University 7 June 2017 Data archiving no no no policy LISA 8 34

7 June 2017 LISA 8 35