Introduction

Background/rationale

Many regulatory efforts to promote data sharing have been made and are underway at the international and national levels for the ultimate purpose of advancing science and research. High-level bodies and funding agencies are the primary contributors to the culture and practice of data sharing [1]. For example, the European Commission implemented the Open Research Data (ORD) pilot in Horizon 2020 (H2020), which requires H2020-funded projects to develop data management plans (DMPs) and provide open access to research data [2]. The principle of ORD is that research data should be “as open as possible, as closed as necessary,” and it emphasizes sound data management rather than forcing all research data to be open [3]. The US National Institute of Health (NIH) released a new Data Management and Sharing (DMS) policy in October 2020, effective as of January 25, 2023, which requires all applicants to submit DMS plans if the proposed research generates scientific data. Similar to the principle of ORD, the new NIH policy intends to “encourage data sharing to the extent that it is possible” [4]. In addition, publishers and individual journals influence researchers’ data sharing behavior by establishing data sharing policies that encourage or require making data available along with the publication of research articles [5,6]. Data sharing and reproduction is one of the policies on publication ethics that the recently updated version of the Principles of Transparency and Best Practice in Scholarly Publishing asks journals to include [7].

Compared to Europe and the United States’ commitment arising from the multiplicity of players in DMS, South Korea remains in the early phases of ORD agenda implementation. A national policy on research data management (RDM) is currently specified in the National Research & Development (R&D) Information Processing Standard. It is an administrative rule enforced by the ordinance of the Ministry of Science and ICT (MSIT) under the National R&D Innovation Act enacted in 2021. However, the policy on RDM is only applied to national R&D projects for which central government agencies consider it necessary to submit DMPs, and data sharing is not even mentioned in the policy [8]. Despite these limitations, government-funded research institutes under the National Research Council of Science and Technology (NST) affiliated with the MSIT have been directly affected by the policy and have gradually adopted DMPs; as of 2021, three of the 25 research institutes under the NST have implemented DMPs and data repositories. These include the Korea Institute of Geoscience and Mineral Resources (KIGAM), the Korea Institute of Oriental Medicine (KIOM), and the Korea Institute of Science and Technology Information (KISTI) [9]. In addition, the Korea Research Institute of Standards and Science (KRISS), another research institute under the NST, operates the National Standard Reference Data Center (NSRDC) [10]. Moreover, the National Institute of Ecology (NIE), a leading government research institute affiliated with the Ministry of Environment, has developed a platform for sharing ecological data named EcoBank [11].

Since the regulatory basis for RDM has been formed in South Korea and corresponding practices have been initiated in several government research institutes, it would be useful to investigate how researchers in the institutes regard data sharing. Such a study would lead to a better understanding of researchers’ experiences and what makes them reluctant when considering data sharing. A few studies have examined perceptions and the status of DMS based on surveys and/or interviews of researchers in Korean government research institutes [12‒14]. However, the existing studies surveyed researchers in a single institute or were conducted before the regulations of RDM were established. Exploring the attitudes toward data sharing of researchers in multiple government research institutes where RDM practices are implemented will help address researchers’ needs regarding their institutes’ data management practices.

Objectives

This study examined the current data sharing practices and perceptions of researchers in five Korean government research institutes currently involved in RDM, which are KIGAM, KIOM, KISTI, KRISS, and NIE. The study assumed that researchers of these institutes have a certain level of understanding about data sharing and thus that it would be appropriate to recruit them as survey participants. The research questions were as follows: (1) To what extent do the researchers share their research data? (2) In what ways do they share research data? (3) What challenges do they perceive in relation to data sharing?

Methods

Ethics statement

This study was exempt from deliberation by the Institutional Review Board because there was no collection of sensitive information or individual identification information.

Study design

This is a descriptive study based on an online survey.

Data collection methods

A survey questionnaire was developed based on studies recently performed regarding data sharing and publication [15‒17]. The questionnaire consisted of 29 questions in four areas: (1) research data creation and management; (2) data sharing and publication; (3) perceptions of data publication; and (4) demographic information. SurveyMonkey was used to construct an online questionnaire, and a link was distributed via a discussion forum on the homepage of each of the five government research institutes from June 15 to 29, 2022. As a result, 224 responses were collected and used for the analysis. Almost all responses were complete, except for eight responses that failed to provide demographic information.

Units of study

The unit of study in this research was the individual, since the survey data were collected from 224 researchers employed in the five Korean government research institutes.

Data analysis

This study focused on analyzing responses concerning the extent, methods, and barriers to data sharing. Descriptive statistical analyses were performed to compare the survey responses in terms of whether and to what extent respondents shared research data, which of the various ways of sharing data they utilized, and the challenges they perceived.

Results

Participants

The demographics of the respondents are presented in Table 1. The majority of respondents were men and had doctoral degrees. Most were in their 30s or 40s. Regarding the disciplines of the respondents, most were in 56 information communications technology (25.9%), 38 biological science (17.6%), 21 public health and medicine (9.7%), and 16 earth science (7.4%). In total, 24 disciplines were reported by the respondents (Fig. 1).

The extent of data sharing

Out of 224 respondents, 180 (80.4%) answered that they had experience collecting or creating research data. Among those 180 researchers, 32 mentioned that they had never shared their research data; thus, the remaining 148 respondents had shared at least some of their data (Table 2). Only seven respondents stated that they shared all of their data, while the majority shared some or most of theirs.

The 148 respondents with data sharing experience were also asked with whom they had shared research data. As shown in Table 3, sharing data with principal investigators or coinvestigators involved in collecting or creating data received the greatest proportion of the responses (n=94, 36.3%), followed by sharing data with all participants in research projects where data were collected/created (n=86, 33.2%). The findings indicate that restricted data sharing—“the exchange of data between the members of a specific group or project consortium to their mutual benefit” [2]—was more common than data sharing on request (n=60, 23.2%) or making data publicly accessible (n=19, 7.3%).

This study also analyzed the responses presented in Table 3 to compare the proportion of respondents who only shared data in a restricted way with those who shared data on request or created an ORD. The responses were collected from a question that allowed multiple answers. Thus, the study identified the number of respondents providing one or two answers who selected only the choices categorized under restricted data sharing. This response type was found to constitute a majority, with a total of 80 such respondents (54.1%) (Table 4). This result indicates that although 148 respondents reported sharing their data, most conducted data exchange with a limited scope (i.e., within a research project) rather than making data available to outside researchers who needed it or to anyone at any time.

Methods of data sharing

In terms of the ways in which the respondents reported sharing their data, 110 responses (44.0%) described sharing data directly to individual researchers on request. Strongly recommended data sharing methods, such as deposition within a data repository, were less likely to be used by the respondents. However, sharing data through an institutional repository or server received the second greatest number of responses (n=45, 18.0%). This might have resulted from the fact that the research institutes with which these respondents are affiliated have implemented data repositories and DMPs. A personal or research team website was the third most used method of sharing data (n=40, 16.0%). Sharing data via journals’ databases or repositories while publishing research articles also received 37 responses (14.8%). This result highlights the role of journals in facilitating data sharing. Only a small portion of responses indicated sharing data via disciplinary repositories or data journals. These results are illustrated in Table 5.

Barriers to data sharing

One survey question asked about factors that made respondents reluctant to share data. This question was asked of the 180 respondents with experience creating or collecting research data, and multiple answers were permitted. As shown in Table 6, the top three reasons received a similar proportion of responses (around 16%). These reasons were the time and effort required to organize data, unclear copyright or ownership of data, and a lack of reward or recognition for sharing data. The factor of concerns about data having sensitive information was also selected in 77 responses (15.3%), followed by concerns about data having errors (n=60, 11.9%). Other options selected by participants included no regulations for data sharing (n=43, 8.5%), the lack of an adequate data sharing platform (n=35, 7.0%), the perceived lack of need for their data (n=23, 4.6%), and the absence of funding for sharing data (n=17, 3.4%) (Table 6).

Discussion

Key results

Among the 148 respondents who had experienced data sharing, the vast majority reported sharing some or most of their data. Sharing data with principal investigators, coinvestigators, or participants in a research project was more common than sharing data with outside researchers on request or creating an ORD. A majority of the 148 respondents with data sharing experience shared data only in a restricted manner. Therefore, the extent of data sharing performed by the respondents is limited in light of international efforts to promote ORD.

The most common method of sharing data was to provide data directly to individual researchers who asked for it, even though funding agencies and journals strongly recommend depositing data in repositories. Yet, institutional repositories or servers were used the second most. This might be because the research institutes with which the respondents were affiliated operated institutional repositories as data sharing platforms. Journals’ databases or repositories were also employed; thus, journals’ role in promoting data sharing should be recognized.

Various factors that impeded data sharing were identified. The most commonly indicated factor was the time and effort required to organize data, followed by concerns about copyright or ownership of data and lack of reward or recognition for data sharing. Respondents also had concerns about sharing data containing sensitive information and the possibility of errors in their data. Exploring ways to mitigate such barriers to data sharing is necessary at an institutional and national level.

Interpretation

Most respondents with data sharing experience made some or most of their data available to others. This result is similar to that of the study by Tenopir et al. [16], which conducted an international survey of researchers’ DMS practices. However, the proportion of respondents in this study who reported sharing all of their data was about four times lower than that of Tenopir et al. [16]. Restricted data sharing was also more common than making data available to outside researchers or everyone. According to the principles of the H2020 ORD pilot and newly released NIH policies, ORD is strongly recommended, but restricted data sharing is reasonable if more value can be provided by restricting data access [2]. Since decisions about access to data are nuanced rather than binary [2], it is important for Korean government research institutes to provide ongoing training and support for affiliated researchers so that they can make informed decisions and adequate justifications if restricted data sharing is needed. In addition, developing an appropriate infrastructure that enables researchers to share and preserve their data is necessary, and funding agencies and journals strongly recommend depositing data into robust repositories [4,5]. Therefore, research institutes should make commitments to implement data repositories and encourage the submission of DMPs to build good DMS practices.

Moreover, a recent systematic review of studies on factors associated with data sharing suggested 11 categories of factors: researcher’s background, requirements and formal obligations, intrinsic motivations, facilitating conditions, trust, expected performance, social influence and affiliation, effort, researcher’s experience, legislation and regulation, and data characteristics [18]. The findings of this study are mostly related to effort (time and effort required to organize data), legislation and regulation (concerns about copyright or ownership and sensitive information), and expected performance (lack of reward and recognition). To alleviate these factors, it is desirable to develop RDM services for researchers, which will reduce time and save effort in data preparation and organization. Furthermore, training and consultation services regarding copyright and privacy will help researchers better understand legal issues and be less concerned about accidentally violating the law when considering data sharing. Finally, as the lack of incentives has been suggested as a major impediment to data sharing [19], continuing efforts and discussions regarding sufficient rewards for data sharing are required within research institutes and externally in disciplines and governments.

Limitations

This study is based on descriptive statistical analyses only. As such, the results show the current state of data sharing behavior and perception reported by respondents, but an inferential interpretation is not possible.

Conclusion

This study found that survey respondents affiliated with Korean government research institutes commonly performed restricted data sharing. ORD is internationally recommended, but at the same time, restricted data sharing is allowable if it is possible to derive value while maintaining restricted access to data. Providing ongoing training and support for researchers can help them make informed decisions about access to data. It is also important for research institutes to develop and sustain institutional data repositories as a platform for DMS. To address barriers to data sharing, it is necessary to implement RDM services that help reduce the effort required to organize data and provide consultations for copyright and privacy issues. It is also necessary to continue exploring possible incentives for researchers who share data from institutional, disciplinary, and governmental perspectives.

Notes

Conflict of Interest

No potential conflict of interest relevant to this article was reported.

Funding

This work was supported by the National Research Council of Science and Technology (No. NST-02).

Fig. 1.

Respondents’ disciplines.

References

• 1. Popkin G. Data sharing and how it can benefit your scientific career. Nature 2019;569:445-7.https://doi.org/10.1038/d41586-019-01506-x. ArticlePubMed
• 2. Austin T, Bei K, Efthymiadis T, Koumoulos EP. Lessons learnt from engineering science projects participating in the Horizon 2020 Open Research Data pilot. Data 2021;6:96. https://doi.org/10.3390/data6090096. Article
• 3. European Commission. H2020 online manual: data management [Internet]. European Union [cited 2023 Jan 21]. Available from: https://ec.europa.eu/research/participants/docs/h2020-funding-guide/cross-cutting-issues/open-access-data-management/data-management_en.htm.
• 4. Gonzales S, Carson MB, Holmes K. Ten simple rules for maximizing the recommendations of the NIH data management and sharing plan. PLoS Comput Biol 2022;18:e1010397. https://doi.org/10.1371/journal.pcbi.1010397. ArticlePubMedPMC
• 5. MacMillan D. Data sharing and discovery: what librarians need to know. J Acad Librariansh 2014;40:541-9.https://doi.org/10.1016/j.acalib.2014.06.011. Article
• 6. Kim Y, Burns CS. Norms of data sharing in biological sciences: the roles of metadata, data repository, and journal and funding requirements. J Inf Sci 2016;42:230-45.https://doi.org/10.1177/0165551515592098. Article
• 7. Committee on Publication Ethics (COPE). Principles of transparency and best practice in scholarly publishing [Internet]. COPE; 2022 [cited 2023 Jan 21]. Available from: https://publicationethics.org/resources/guidelines/principles-transparency-and-best-practice-scholarly-publishing.
• 8. Shmagun H, Shim J, Choi KN, Shin SK, Kim J, Oppenheim C. Korea’s national approach to Open Science: present and possible future. J Inf Sci 2022 Jul 12 [Epub]. https://doi.org/10.1177/01655515221107336. Article
• 9. Kim J, Kim S. A study on the development of the template for creating the data management plans. J Korean Soc Lib Inf Sci 2021;55:373-96.https://doi.org/10.4275/KSLIS.2021.55.3.373. Article
• 10. National Center for Standard Reference Data (NCSRD). Introduction to data center [Internet]. NCSRD; c2022 [cited 2023 Jan 21]. https://www.srd.re.kr/css/dcintro/S_CSS_030100.do?menuId=M_CSS_03_01.
• 11. Kim HW, Yoon S, Kim M, Shin M, Yoon H, Kim K. EcoBank: a flexible database platform for sharing ecological data. Biodivers Data J 2021;9:e61866. https://doi.org/10.3897/BDJ.9.e61866. ArticlePubMedPMC
• 12. Kim M, Kim S. A study on the factors affecting sharing of research data of science and technology researchers. J Korean Soc Lib Inf Sci 2015;49:313-34.https://doi.org/10.4275/KSLIS.2015.49.2.313. Article
• 13. Choi M, Lee SB, Lee S. Research data management of science and technology research institutes in Korea. J Korea Contents Assoc 2017;17:117-26.https://doi.org/10.5392/JKCA.2017.17.12.117. Article
• 14. Park Y, Kim J. A study on research data creation and management by researchers in mechanical engineering. J Korean Soc Arch Rec Manag 2021;21:137-62.https://doi.org/10.14404/JKSARM.2021.21.4.137. Article
• 15. Korea Social Science Data Archive (KOSSDA). A survey of researchers about research data creation and management. KOSSDA; 2021.
• 16. Tenopir C, Dalton ED, Allard S, et al. Changes in data sharing and data reuse practices and perceptions among scientists worldwide. PLoS One 2015;10:e0134826. https://doi.org/10.1371/journal.pone.0134826. ArticlePubMedPMC
• 17. Kratz JE, Strasser C. Researcher perspectives on publication and peer review of data. PLoS One 2015;10:e0117619. https://doi.org/10.1371/journal.pone.0117619. ArticlePubMedPMC
• 18. Zuiderwijk A, Shinde R, Jeng W. What drives and inhibits researchers to share and use open research data? A systematic literature review to analyze factors influencing open research data adoption. PLoS One 2020;15:e0239283. https://doi.org/10.1371/journal.pone.0239283. ArticlePubMedPMC
• 19. Devriendt T, Shabani M, Lekadir K, Borry P. Data sharing platforms: instruments to inform and shape science policy on data sharing? Scientometrics 2022;3007-19.https://doi.org/10.1007/s11192-022-04361-2. Article

Figure & Data

References

Citations

Citations to this article as recorded by  

• Korean scholarly journal editors’ and publishers’ attitudes towards journal data sharing policies and data papers (2023): a survey-based descriptive study
  Hyun Jun Yi, Youngim Jung, Hyekyong Hwang, Sung-Nam Cho
  Science Editing.2023; 10(2): 141.     CrossRef
• Data sharing and data governance in sub-Saharan Africa: Perspectives from researchers and scientists engaged in data-intensive research
  Siti M. Kabanda, Nezerith Cengiz, Kanshukan Rajaratnam, Bruce W. Watson, Qunita Brown, Tonya M. Esterhuizen, Keymanthri Moodley
  South African Journal of Science.2023;[Epub]     CrossRef
• Identifying key factors and actions: Initial steps in the Open Science Policy Design and Implementation Process
  Hanna Shmagun, Jangsup Shim, Jaesoo Kim, Kwang-Nam Choi, Charles Oppenheim
  Journal of Information Science.2023;[Epub]     CrossRef