Accountability in Research-Policies and Quality Assurance最新文献

Research Funding Organizations (RFOs) play an important role in promoting research integrity (RI). Not only do they allocate resources to research institutions and researchers, but they also set and monitor research standards in their funded projects. In spite of their crucial role, there is a paucity of guidance on how RFOs can promote research integrity. As part of the EU-Funded SOPs4RI project, we aimed to address this gap by co-creating guidelines to help RFOs promote RI, engaging a diverse group of stakeholders. Based on a Delphi survey, reviews of evidence and stakeholder interviews, three guideline topics were identified: 1) the selection and evaluation of proposals; 2) monitoring of funded projects; and 3) prevention of unjustified interference. Four sets of co-creation workshops were conducted for each guideline topic, and the input revised and finalized. Understanding these debates could help RFOs from diverse cultural and organizational backgrounds who are developing their own RI guidelines. Therefore, in this paper, we summarize the key results and emphasize the final recommendations. Further, we provide the main points of discussion that occurred during the workshops and explain how they were addressed or resolved in the final guidelines and how they can help in future endeavors to improve funders' practices to foster RI.

Background: This study explored experiences of quantitative researchers who work with marginalized populations.Methods/materials: Participants were recruited from the Society for Personality and Social Psychology forum, and asked questions regarding their struggles and best practices while working with marginalized populations in which they are or are not a member.Results: Responses included concerns regarding bidirectional trust, community norms, perceived bias, diversity and participant recruitment and compensation. We explore the benefits of qualitative understandings of bias (i.e. positionality, reflexivity), salient concerns reported by quantitative researchers, and our recommendations for the ethical inclusion of these practices across quantitative work.Conclusions: This paper contributes to understanding of current struggles and best practices while conducting research among marginalized populations. Additionally, we encourage quantitative researchers to engage in reflexive research practices, particularly for the benefit of marginalized group research. We extend the insider-outsider researcher discussion to quantitative researchers.

Background: Understanding review comments holds significant importance within the realm of scientific discourse. This study aims to conduct an empirical analysis of factors associated with praise in peer review.Methods: The study involved manual labeling of "praise" in 952 review comments drawn from 301 articles published in the British Medical Journal, followed by regression analysis.Results: The study reveals that authors tend to receive longer praise when they share a cultural proximity with the reviewers. Additionally, it is observed that female reviewers are more inclined to provide praiseConclusions: In summary, these discoveries contribute valuable insights for the development of a constructive peer review process and the establishment of a more inclusive research culture.

The recent emergence of Large Language Models (LLMs) and other forms of Artificial Intelligence (AI) has led people to wonder whether they could act as an author on a scientific paper. This paper argues that AI systems should not be included on the author by-line. We agree with current commentators that LLMs are incapable of taking responsibility for their work and thus do not meet current authorship guidelines. We identify other problems with responsibility and authorship. In addition, the problems go deeper as AI tools also do not write in a meaningful sense nor do they have persistent identities. From a broader publication ethics perspective, adopting AI authorship would have detrimental effects on an already overly competitive and stressed publishing ecosystem. Deterrence is possible as backward-looking tools will likely be able to identify past AI usage. Finally, we question the value of using AI to produce more research simply for publication's sake.

The founders of PubPeer envisioned their website as an online form of a "journal club" that would facilitate post-publication peer review. Recently, PubPeer comments have led to a significant number of research misconduct proceedings - a development that could not have been anticipated when the current federal research misconduct regulations were developed two decades ago. Yet the number, frequency, and velocity of PubPeer comments identifying data integrity concerns, and institutional and government practices that treat all such comments as potential research misconduct allegations, have overwhelmed institutions and threaten to divert attention and resources away from other research integrity initiatives. Recent, high profile research misconduct cases accentuate the increasing public interest in research integrity and make it inevitable that the use of platforms such as PubPeer to challenge research findings will intensify. This article examines the origins of PubPeer and its central role in the modern era of online-based scouring of scientific publications for potential problems and outlines the challenges that institutions must manage in addressing issues identified on PubPeer. In conclusion, we discuss some potential enhancements to the investigatory process specified under federal regulations that could, if implemented, allow institutions to manage some of these challenges more efficiently.

Background: The rapid acceleration of authorship inflation-increasing numbers of authors per publication in collaborative research-has rendered the traditional "substantial contributions" criterion for authorship and the lack of transparency in author contributions increasingly problematic.Methods and results: To address these challenges, a revamped approach to authorship is proposed, replacing the rigid requirement of "substantial contributions" with a more flexible, project-specific criterion of "sufficient contributions," as determined and justified by the authors for each project. This change more accurately reflects and accommodates the proliferation of scientific collaboration ("team science" or "group science"). It broadens the scope and granularity of roles deserving of authorship by integrating the Contributor Roles Taxonomy (CRediT) and Method Reporting with Initials for Transparency (MeRIT) systems. It mandates in-text documentation of who did what (e.g., who collected what data) and moves beyond the typical binary (all-or-none) classification by assigning a gradated contribution level to each author for each role. Contributions can be denoted using an ordinal scale-either coarse (e.g., lead, equal, and supporting) or fine-grained (e.g., minimal, slight, moderate, substantial, extensive, and full). To support the implementation of the revamped approach, an authorship policy template is provided.Conclusions: Adopting proportional, role-specific credit allocation and explicit documentation of contributions fosters a more transparent, equitable, and trustworthy scientific environment.

Purpose: The increasing use of AI tools, particularly large language models like ChatGPT, in academic research has raised significant questions about authorship and transparency. This commentary emphasizes the need for a standardized AI contributions taxonomy to clarify AI's role in producing and publishing research outputs, ensuring ethical standards and maintaining academic integrity.

Approach: We propose adapting the NIST AI Use Taxonomy and incorporating categories that reflect AI's use in tasks such as hypothesis generation, data analysis, manuscript preparation, and ethical oversight. Findings: Establishing an AI contributions taxonomy for the production and publication of research output would address inconsistencies in AI disclosure, enhance transparency, and uphold accountability in research. It would help differentiate between AI-assisted and human-led tasks, providing more explicit attribution of contributions.

Findings: Establishing an AI contributions taxonomy for the production and publication of research output would address inconsistencies in AI disclosure, enhance transparency, and uphold accountability in research. It would help differentiate between AI-assisted and human-led tasks, providing more explicit attribution of contributions.

Practical implications: The proposed taxonomy would offer researchers and journals a standardized method for disclosing AI's role in academic work, promoting responsible and transparent reporting aligned with ethical guidelines from COPE and ICMJE.

Value: A well-defined AI contributions taxonomy for the production and publication of research output would foster transparency and trust in using AI in research, ensuring that AI's role is appropriately acknowledged while preserving academic integrity.

Background: This autoethnographic study examines email invitations for health researchers to publish in journals outside their expertise, exploring implications for interdisciplinary research and knowledge production.Methods: Over three months, email invitations to publish outside the author's field were documented and analysed thematically and through reflexive journaling.Results: Five main themes in publication invitations were identified: emphasising novelty, promising rapid publication, appealing to research impact, flattering language, and persistent messaging. Reflexive analysis revealed complex factors shaping responses, including publication pressures, desires for prestige, and tensions between disciplinary norms and interdisciplinary collaboration. While invitations may present opportunities for novel collaborations, they often reflect predatory publishing practices.Conclusions: Navigating this landscape requires careful discernment, commitment to academic integrity, and reflexivity about one's positionality. The study underscores the need for researchers to critically interrogate the motivations behind such invitations. Further research could explore decision-making processes across disciplines and implications for academic publishing integrity and equity.

Background: Effective leadership and management practices contribute to responsible, high-quality research and the well-being of team members. We describe the development and initial validation of a measure assessing principal investigators' leadership and management practices and a measure of research team practices.

Methods: Using a cross-sectional survey design, 570 postdoctoral researchers funded by the National Institutes of Health reported on the perceived behaviors of their principal investigator (PI) and the research team. The PI leadership and management items factored into two dimensions: fostering relationships and directing research.

Results: Correlations of these new scales with existing, validated measures of ethical leadership and general leader behavior provided evidence of convergent validity. Providing evidence for criterion-related validity, scores on the new measures predicted lab climate for research ethics, self-reported productivity, and job satisfaction. Research team practices provided additional predictive value beyond leadership and management behaviors.

Conclusions: This study provides construct validity evidence for the new Leadership and Management in Science (LAMPS) Measure and the Research Team Practices (RTP) Measure. Qualitative responses to an open-ended item reinforced the importance of relationships and directive supervision for a positive environment. These measures can be useful tools for future research or may be useful for PIs seeking feedback about their practices.

Background: Scientific misinformation is a much-discussed topic, and the COVID-19 crisis has highlighted the importance of reliability in science and research. However, limiting misinformation is complicated because of the growing number of communication channels, in which scientific and nonscientific content are often mixed.

Methods: This case study combines the examination of references, online observation, and a content and frequency analysis to investigate the dissemination of scientific misinformation in the interplay of different genres and media.

Results: Using the example of the claimed existence of 34,000 human emotions, this study demonstrates how questionable statements are spread in science, popular science, and pseudoscience, making it particularly challenging to track and correct them.

Conclusions: The findings highlight epistemic authority, trust, and injustice within and between scientific and nonscientific communities. The author argues that, in the digital age, researchers should defend and monitor scientific principles beyond academia.