Community developing and supporting practices to increase replicability of scient

社区开发和支持实践以提高科学的可复制性

• 批准号: 8756974
• 负责人: Brian A Nosek
• 金额: $ 21.51万
• 依托单位: CENTER FOR OPEN SCIENCE
• 依托单位国家: 美国
• 财政年份: 2014
• 资助国家: 美国
• 起止时间: 2014-09-15 至 2019-04-30
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/8756974
• 关键词: Address; Adopted; Adoption; Aging; Attention; Behavior; Communities; Community Practice; Complex; Data; Data Analyses; Data Collection; Disclosure; Education; Goals; Heart; Incentives; Journals; Knowledge; Measures; Peer Review; Publishing; Reporting; Research; Research Design; Research Personnel; Rewards; Role; Sample Size; Science; Scientist; Solutions; Structure; Testing; Time; base; behavioral/social science; cost; experience; flexibility; forgetting; improved; innovation; insight; public health relevance; tool

DESCRIPTION (provided by applicant): Replicability is a core value of science to assure that the knowledge is valid. However, the culture of scientific incentives and rewards prioritizes innovation almost exclusively at the expense of replication and verification. This project supports developing a community to shift incentives so that replicability is rewarded in aging research and social-behavioral sciences more generally. The community will create, evaluate, and disseminate four projects that could shift incentives to improve replicability. The four projects are: (1) Badges to acknowledge and reward open practices: Currently, scientists earn reputation and reward for publishing their research in journals. Replicability is enhanced if other scientists can access the materials and data that were the basis for the published research. The proposed badges will attach to articles to reward authors for making their research openly available to others. (2) Disclosure standards: The replicability of research is partly contingent on how the evidence was obtained. There are some things that every report should disclose. For example: how the sample size was determined, whether there were other measures not reported, and whether the results are dependent on other measures used in the analysis. (3) Registered reports: Researchers do not conduct replications because they are hard to publish. Registered reports will introduce a publishing format in which the study design is reviewed before the data are collected. This is good for replication research because researchers can find out before the effort of data collection if the study will be publishable. (4) Checklists for research workflow: Science is complex. It is easy to forget or fail to include important information in a report, or t check for particular things when being a reviewer or editor. Checklists can be useful guides for making sure particular practices are followed.

描述(申请人提供)：可复制性是科学的核心价值，以确保知识的有效性。然而，科学激励和奖励的文化将创新放在首位，几乎完全以复制和验证为代价。该项目支持 开发一个社区来改变激励机制，以便在老龄化研究和更广泛的社会行为科学中奖励可复制性。该社区将创建、评估和传播四个项目，这些项目可以改变激励措施，以提高可复制性。这四个项目是：(1)表彰和奖励开放实践的徽章：目前，科学家通过在期刊上发表他们的研究而获得声誉和奖励。如果其他科学家 可以访问作为已发表研究的基础的材料和数据。拟议的徽章将被贴在文章上，以奖励将他们的研究公开给其他人的作者。(2)披露标准：研究的可重复性在一定程度上取决于证据是如何获得的。有些事情是每一份报告都应该披露的。例如：样本量是如何确定的，是否有其他未报告的指标，以及结果是否取决于分析中使用的其他指标。(3)注册报告：研究人员不进行复制，因为它们很难发表。注册报告将引入一种发布格式，在收集数据之前对研究设计进行审查。这对复制研究是有利的，因为研究人员可以在数据收集之前发现这项研究是否可以发表。(4)研究工作流程清单：科学是复杂的。作为审查者或编辑，很容易忘记或未能在报告中包含重要信息，或者不检查特定的内容。核对表是确保特定做法得到遵守的有用指南。