Dies ist eine Übersichtsseite mit Metadaten zu dieser wissenschaftlichen Arbeit. Der vollständige Artikel ist beim Verlag verfügbar.
Development of a checklist to assess and classify initiatives that foster responsible research practices - results from the INSPIRE project
0
Zitationen
4
Autoren
2020
Jahr
Abstract
Background<br> In recent years several Codes of Conduct for Research Integrity were developed or revised. However, having a code does not imply that all stakeholders are aware of the current standards, let alone live up to them.<br> The INSPIRE project (Inventory in the Netherlands of Stakeholders’ Practices and Initiatives on Research integrity to set an Example)1 aims to collect, classify and share initiatives to foster research integrity, with a view to inspire and enable. A checklist is developed to:<br> - Structure a toolbox2 of initiatives and the information included;<br> - Assess initiatives on feasibility and usefulness;<br> - Assist initiators in the development of initiatives.<br> Methods<br> A 3-round Delphi study was used to reach consensus on the checklist. A Delphi is a structured group process that allows for individual differences and changes of views, to ultimately deal with a complex issue.<br> The first round aimed to identify relevant characteristics of the checklist, relevant items for the taxonomy of initiatives in the toolbox, and indicators of quality, feasibility and suitability of initiatives. Round two focussed on operationalisation of the taxonomy and the abovementioned indicators.<br> The first draft of the checklist was piloted in a workshop at the 6th World Conference on Research Integrity. This provided additional input for the development of the checklist. Subsequently INSPIRE’s advisory group gave feedback on the renewed draft. A third Delphi round is planned with the purpose of finetuning the checklist.<br> Results<br> At the time of writing this abstract, the second round of the Delphi study is finished. The checklist will be finalised after the third round and presented at the REWARD|EQUATOR conference. The current draft consists two main assessment items (potential effectiveness; potential to be implemented), and the classification is split in four classifiers (type; audience; participation level; topic) and a reporting guideline of 15 categories and six open ended topics.<br> Conclusions<br> The development of a checklist that is applicable to a broad spectrum of initiatives and stakeholders is challenging. The Delphi and the workshop have yielded plenty of valuable input and shaped the checklist into an instrument that is supported by a wide audience.<br> References (max 2)<br> 1. Netherlands Research Integrity Network. Inventory in the Netherlands of Stakeholders’ Practices and Initiatives on Research integrity to set an Example (INSPIRE project) https://www.nrin.nl/about/inspire-project/<br> 2. The inventory of initiatives and composition of the toolbox are presented in another abstract that is submitted to this conference by DF van der Schot et al.
Ähnliche Arbeiten
Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI
2019 · 8.260 Zit.
Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead
2019 · 8.116 Zit.
High-performance medicine: the convergence of human and artificial intelligence
2018 · 7.493 Zit.
Proceedings of the 19th International Joint Conference on Artificial Intelligence
2005 · 5.776 Zit.
Peeking Inside the Black-Box: A Survey on Explainable Artificial Intelligence (XAI)
2018 · 5.438 Zit.