Component-Based Research (CBR) in STEM Education: Exploring a Potentially Transformative Research Paradigm and Infrastructure

本次中型研究基础设施会议将探讨基于组件的研究（CBR）作为STEM教育研究的可能新范式。此外，与会者将探讨什么可能构成一个相应的基础设施，以支持社区康复的广泛使用。在使用CBR时，研究人员不仅将教育创新作为一个整体，而且还将其单独的部分。此外，它们还系统地描述了创新发生的背景。通过这样做，研究人员可以确定的不仅仅是一个程序是否“有效”。随着时间的推移，CBR可能使研究人员能够识别在特定环境中适用于特定学生群体的创新部分，从而产生可定制的和更具影响力的创新。CBR范式也可能有助于解决STEM教育研究中一些长期存在的挑战。例如，虽然许多研究为许多教育改进的实例提供了信息，但随着时间的推移，该领域在积累知识方面并不成功。基于实例的推理对创新的具体描述和环境因素可能有助于解决这一问题。此外，在一种情况下运作良好的STEM教育项目不一定在另一种情况下运作良好，这给当地教育工作者带来了在没有指导的情况下适应创新的挑战。基于实例的推理方法，沿着支持研究基础设施，可以支持基于数据的适应，使从业者能够根据他们的具体情况定制教育创新。CBR方法也有可能使数据更容易提供给从业者，教学材料开发人员和其他可以使用它来通知他们的工作的人。这个为期两天的会议将汇集30个人来检查和阐明将CBR范式带到STEM教育研究的广泛领域的可能步骤。与会者将包括在STEM教育和STEM教育研究领域从事各种工作的个人，以及在公共卫生等其他学科中使用类似CBR方法的学者。此外，与会者将包括来自历史上被排除在STEM领域之外的群体的代表。会议的两个主要成果是1）社区建设和知识生成，以及2）总结会议工作和后续步骤的专著。本次会议将重点关注该领域如何在STEM教育研究中实施CBR范式。它将按需要发展的五个方面组织，以便为进一步的社区康复工作奠定基础，并额外关注这项工作所需的研究基础设施。具体来说，将有五个工作组研究：1）开发协作语言的方法，包括分类法，本体论和标准; 2）可以解决CBR分析复杂性的创新分析方法; 3）CBR如何利用数据挖掘，人工智能和其他支持CBR的数据科学元素; 4）数据库，设计和其他基础设施，以确保基于CBR的数据的可访问性，并支持其在该领域的应用;以及5）参与该领域，发展合作和共识，以及创建培训和继续教育的挑战。与小组工作交替进行的将是“协商研讨会”，这意味着工作组向其他工作组介绍他们的工作，其他工作组将对工作进行询问，以提出挑战，获得新的见解，并确定要发展的想法以及要放弃的想法。会议结束后，将邀请与会者通过撰写文稿或通过审查为撰写专著作出贡献。这个项目是通过与比尔梅林达盖茨基金会，施密特期货，沃尔顿家族基金会的合作伙伴关系支持。该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

This Mid-scale Research Infrastructure Conference will explore component-based research (CBR) as a possible new paradigm for research in STEM education. In addition, participants will explore what might comprise a corresponding infrastructure to support the widespread use of CBR. When using CBR, researchers look not only at an educational innovation as a whole, but also at its individual parts. Additionally, they systematically describe the contexts within which the innovation is happening. By doing so, researchers can determine more than whether a program “works.” Over time, CBR may enable researchers to identify the parts of an innovation that work for particular groups of students in particular settings, resulting in customizable and more impactful innovations. A CBR paradigm may also help address some long-standing challenges in STEM education research. For example, while much research has informed many instances of educational improvement, the field has been less successful at accumulating knowledge over time. CBR, with its specific descriptions of innovation and circumstantial components may assist with this problem. Also, STEM education programs that work well in one situation don’t necessarily work well in a different one, leaving local educators the challenge of adapting the innovation without guidance. CBR approaches, along with supporting research infrastructure, can support data-informed adaptations that enable practitioners to customize education innovations for their specific circumstances. CBR approaches also have potential to make data more readily available to practitioners, instructional materials developers, and others who can use it to inform their work. This two-day conference will bring together 30 individuals to examine and articulate possible steps for bringing the CBR paradigm to the broad field of STEM education research. Attendees will include individuals who work in a range of capacities in STEM education and STEM education research, as well as scholars who have been using CBR-like approaches in other disciplines such as public health. Additionally, attendees will include a strong representation of people from groups who have historically been excluded from STEM fields. The two primary outcomes of the meeting are 1) community building and knowledge generation, and 2) a monograph summarizing the work of the conference and next steps. This conference will focus on how the field might operationalize the CBR paradigm in STEM education research. It will be organized by five dimensions that need to be developed in order to have a foundation for further CBR work, with additional attention on the research infrastructure needed for such work. Specifically, there will be five working groups that examine: 1) approaches to developing collaborative language, including taxonomies, ontologies, and standards; 2) innovative analytic approaches that can address the complexity of CBR analyses; 3) ways CBR can leverage data mining, AI, and other elements of data science that can support CBR; 4) databases, design, and other infrastructure to ensure accessibility to CBR-based data and support its application in the field; and 5) the challenge of engaging the field, developing collaborations and consensus, and creating training and further education. Alternating with group work will be “consultation seminars” that entail having working groups present their work to other working groups who will interrogate the work to raise challenges, gain new insights, and identify ideas to develop as well as those to jettison. After the conference, participants, will be invited to contribute to the creation of a monograph either through writing contributions or through review. This project is supported through a partnership with the Bill & Melinda Gates Foundation, Schmidt Futures, and the Walton Family Foundation. Funding is also provided by NSF’s EDU Core Research Program.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.