Étiquette : research data

Publié le

Données ouvertes liées et recherche historique : un changement de paradigme

Auteur/Author : Francesco Beretta

Dans le contexte de la transition numérique, le Web sémantique et les données ouvertes liées (linked open data [LOD], en anglais) jouent un rôle de plus en plus central, car ils permettent de construire des « graphes d’information » (knowledge graphs, en anglais) reliant l’ensemble des ressources du Web.

Ce phénomène interroge les sciences historiques et soulève la question d’un changement de paradigme. Après avoir précisé ce qu’il faut entendre par « données », l’article analyse la place qu’elles occupent dans le processus de production du savoir.

Il présente les principales composantes du changement de paradigme, en particulier le potentiel des LOD et d’une sémantique robuste en tant que véhicules d’une information factuelle de qualité, intelligible et réutilisable. S’ensuit une présentation des projets d’infrastructure réalisés au sein du Laboratoire de recherche historique Rhône-Alpes (Larhra) : symogih.org, ontome.net, geovistory.org.

Leur but est de faciliter la transition numérique grâce à un outillage construit en cohérence avec l’épistémologie des sciences historiques et de contribuer à la réalisation d’un « graphe d’information » disciplinaire.

URL : Données ouvertes liées et recherche historique : un changement de paradigme

DOI : https://doi.org/10.4000/revuehn.3349

Publié le

It Takes a Researcher to Know a Researcher: Academic Librarian Perspectives Regarding Skills and Training for Research Data Support in Canada

Author : Alisa B. Rod

Objective

This empirical study aims to contribute qualitative evidence on the perspectives of data-related librarians regarding the necessary skills, education, and training for these roles in the context of Canadian academic libraries.

A second aim of this study is to understand the perspectives of data-related librarians regarding the specific role of the MLIS in providing relevant training and education. The definition of a data-related librarian in this study includes any librarian or professional who has a conventional title related to a field of data librarianship (i.e., research data management, data services, GIS, data visualization, data science) or any other librarian or professional whose duties include providing data-related services within an academic institution.

Methods

This study incorporates in-depth qualitative empirical evidence in the form of 12 semi-structured interviews of data-related librarians to investigate first-hand perspectives on the necessary skills required for such positions and the mechanisms for acquiring and maintaining such skills.

Results

The interviews identified four major themes related to the skills required for library-related data services positions, including the perceived importance of experience conducting original research, proficiency in computational coding and quantitative methods, MLIS-related skills such as understanding metadata, and the ability to learn new skills quickly on the job.

Overall, the implication of this study regarding the training from MLIS programs concerning data-related librarianship is that although expertise in metadata, documentation, and information management are vital skills for data-related librarians, the MLIS is increasingly less competitive compared with degree programs that offer a greater emphasis on practical experience working with different types of data in a research context and implementing a variety of methodological approaches.

Conclusion

This study demonstrates that an in-depth qualitative portrait of data-related librarians within a national academic ecosystem provides valuable new insights regarding the perceived importance of conducting original empirical research to succeed in these roles.

URL : It Takes a Researcher to Know a Researcher: Academic Librarian Perspectives Regarding Skills and Training for Research Data Support in Canada

DOI : https://doi.org/10.18438/eblip30297

Publié le

Modèles et outils pour la publication de métadonnées d’archives géographiques et de leurs données dérivées

Auteur.ices/Authors : Hersent Melvin, Abadie Nathalie, Duménieu Bertrand, Perret Julien

L’interopérabilité des données dans un projet pluridisciplinaire est primordiale. Prenant l’exemple d’un projet de recherche en histoire spatiale, nous comparerons dans un premier temps les standards et vocabulaires à notre disposition pour décrire des données géographiques et des documents d’archives.

Nous proposons ensuite un alignement entre les standards retenus : l’ISO 19115 et RiC-O. Enfin, nous proposons une architecture de microservices pour la saisie, le stockage, la publication sur le Web et l’interrogation unifiée des métadonnées de nos sources.

URL : Modèles et outils pour la publication de métadonnées d’archives géographiques et de leurs données dérivées

Original location : https://hal.science/hal-04110787

Publié le

La crédibilité des matériaux ethnographiques face au mouvement d’ouverture des données de la recherche

Auteur.ices/Authors : Alix Levain, Florence Revelin, Anne-Gaëlle Beurier, Marianne Noël

Les politiques d’ouverture des données de la recherche s’appuient sur des arguments de transparence, d’innovation et de démocratisation des savoirs. Cet article vise à rendre intelligibles leurs implications pour les communautés travaillant à partir de données ethnographiques, confrontées à une transformation des critères de reconnaissance de la crédibilité des savoirs qu’elles produisent.

Alors que les chercheur·e·s qui pratiquent l’ethnographie sont engagé·e·s dans des formes situées de partage des matériaux avec les pair·e·s, les autres disciplines et les « communautés sources », le renforcement du contrôle externe sur les conditions dans lesquelles ce partage s’effectue déstabilise les économies de la crédibilité qui structurent ces pratiques.

Davantage qu’une réticence au processus d’ouverture, le retrait des ethnographes du mouvement apparaît au terme de notre analyse comme résultant à la fois de l’existence d’écologies alternatives des matériaux empiriques et d’une éthique des marges incorporée dans des normes professionnelles souvent implicites.

DOI : https://doi.org/10.4000/rac.30291

Publié le

Biomedical supervisors’ role modeling of open science practices

Authors : Tamarinde L Haven, Susan Abunijela, Nicole Hildebrand

Supervision is one important way to socialize Ph.D. candidates into open and responsible research. We hypothesized that one should be more likely to identify open science practices (here publishing open access and sharing data) in empirical publications that were part of a Ph.D. thesis when the Ph.D. candidates’ supervisors engaged in these practices compared to those whose supervisors did not or less often did.

Departing from thesis repositories at four Dutch University Medical centers, we included 211 pairs of supervisors and Ph.D. candidates, resulting in a sample of 2062 publications. We determined open access status using UnpaywallR and Open Data using Oddpub, where we also manually screened publications with potential open data statements. Eighty-three percent of our sample was published openly, and 9% had open data statements.

Having a supervisor who published open access more often than the national average was associated with an odds of 1.99 to publish open access. However, this effect became nonsignificant when correcting for institutions. Having a supervisor who shared data was associated with 2.22 (CI:1.19–4.12) times the odds to share data compared to having a supervisor that did not.

This odds ratio increased to 4.6 (CI:1.86–11.35) after removing false positives. The prevalence of open data in our sample was comparable to international studies; open access rates were higher. Whilst Ph.D. candidates spearhead initiatives to promote open science, this study adds value by investigating the role of supervisors in promoting open science.

URL : Biomedical supervisors’ role modeling of open science practices

DOI : https://doi.org/10.7554/eLife.83484

Publié le

“We Share All Data with Each Other”: Data-Sharing in Peer-to-Peer Relationships

Author : Eva Barlösius

Although the topic of data-sharing has boomed in the past few years, practices of datasharing have attracted only scant attention within working groups and scientific cooperation (peer-to-peer data-sharing).

To understand these practices, the author draws on Max Weber’s concept of social relationship, conceptualizing data-sharing as social action that takes place within a social relationship. The empirical material consists of interviews with 34 researchers representing five disciplines—linguistics, biology, psychology, computer sciences, and neurosciences.

The analysis identifies three social forms of data-sharing in peer-to-peer relationships: (a) closed communal sharing, which is based on a feeling of belonging together; (b) closed associative sharing, in which the participants act on the basis of an agreement; and (c) open associative sharing, which is oriented to “institutional imperatives” (Merton) and to formal regulations.

The study shows that far more data-sharing is occurring in scientific practice than seems to be apparent from a concept of open data alone. If the main goal of open-data policy programs is to encourage researchers to increase access to their data, it could be instructive to study the three forms of data-sharing to improve the understanding of why and how scientists make their data accessible to other researchers.

URL : “We Share All Data with Each Other”: Data-Sharing in Peer- to-Peer Relationships

DOI : https://doi.org/10.1007/s11024-023-09487-y

Publié le

Initial insight into three modes of data sharing: Prevalence of primary reuse, data integration and dataset release in research articles

Authors : Yukiko Sakai, Yosuke Miyata, Keiko Yokoi, Yuqing Wang, Keiko Kurata

While data sharing has received research interest in recent times, its real status remains unclear, owing to its ambiguous concept. To understand the current status of data sharing, this study examined primary reuse, data integration, and dataset release as the actual practices of data sharing.

A total of 963 articles, chosen from those published in 2018 and registered in the Web of Science global citation database, were manually checked. Existing data were reused in the mode of data integration (13.3%) as frequently as they were for the mode of primary reuse (12.1%). Dataset release was the least common mode (9.0%).

The results show the variation in data sharing and indicate the need for standardization of data description in articles based on thorough registration and expansion in public data archives to close the loop that results in the virtuous cycle of research data.

URL : Initial insight into three modes of data sharing: Prevalence of primary reuse, data integration and dataset release in research articles

DOI : https://doi.org/10.1002/leap.1546