Collaborative Research: Assessing "Systems Thinking" Skills and Learning in Interdisciplinary STEM Courses

这是一个由密歇根州立大学（奖励DUE-1711260）、美国自然历史博物馆（奖励DUE-1711411）和罗格斯大学-新不伦瑞克分校（奖励DUE-1712034）的研究人员参与的合作项目。理解复杂的跨学科问题，如大规模生态系统变化、粮食安全和全球渔业衰退，需要下一代科学家发展系统思维（ST）技能，使他们能够理解人类和自然系统耦合的结构和功能，并确定最有效和最有效地解决问题的杠杆点。能够深入思考系统的复杂动态的学生能够更好地预测系统的行为，设计更有利的结果，并评估不同政策和管理决策之间的权衡。然而，尽管科技技能通常是全国大学可持续性科学课程的基础，但可供教师在课堂上促进和评估科技技能的资源很少。在这个项目中，研究人员将测试一种在STEM课堂中使用认知映射软件Mental Modeler （http://www.mentalmodeler.org/）来教学、评估和提高ST技能的方法，该软件允许学生通过易于使用的系统建模工具来表达他们对问题的理解，并重新排列问题的组成部分来分析场景并修改他们的理解。这种方法将在几个环境科学、环境研究和与可持续性相关的课程中进行测试，这些课程的重点是与粮食系统相关的问题。在之前nsf资助的研究的基础上，该项目将(1)生产支持和测量STEM综合学科中ST学习的工具和模块；(2)进一步开发和测试用于可视化复杂系统的Mental Modeler软件。Mental Modeler是一个多维可视化和建模软件平台，它允许用户识别系统的结构、功能和杠杆点，并运行系统级场景分析。该软件将帮助教育工作者确定适用于STEM领域的科技评估方面，并衡量学生在一段时间内的科技学习情况。具体而言，该项目将解决三个问题：(1)适用于STEM领域的标准化ST评估维度是什么？(2)学生的科技学习进展是否可以通过结构网络指标、功能场景分析和半定量认知图(a)中定义的定性描述(b)来衡量？(3)学生在科技方面是否有可预测的学习轨迹或进展？

This is a collaborative project involving investigators from Michigan State University (Award DUE-1711260), the American Museum of Natural History (Award DUE-1711411), and Rutgers University-New Brunswick (Award DUE-1712034). Understanding complex interdisciplinary issues such as large-scale ecosystem change, food security, and the decline in global fisheries requires that the next generation of scientists develop Systems Thinking (ST) skills that will allow them to understand the structure and function of coupled human and natural systems and identify the leverage points for addressing problems most effectively and efficiently. Students who are able to think deeply about the complex dynamics of a system are better prepared to predict the system's behavior, engineer more favorable outcomes, and evaluate the trade-offs between different policy and management decisions. However, even though ST skills are often the foundation of sustainability science curricula at universities nationwide, few resources are available for instructors to promote and assess ST skills in their classrooms. In this project, the investigators will test a method for teaching, assessing, and improving ST skills in STEM classrooms using cognitive mapping software called Mental Modeler (http://www.mentalmodeler.org/), which allows students to represent their understanding of a problem through easy-to-use system-modeling tools and to rearrange components of the problem to analyze scenarios and revise their understanding. This approach will be tested in several environmental science, environmental studies, and sustainability-related courses focusing on problems related to food systems.Building on previous NSF-funded research, this project will (1) produce tools and modules that support and measure ST learning in integrated STEM disciplines and (2) further develop and test the Mental Modeler software for visualizing complex systems. Mental Modeler is a multidimensional visualization and modeling software platform that allows users to identify the structure, function, and leverage points of a system and to run system-level scenario analyses. The software will help educators determine aspects of ST assessment that are applicable across STEM fields and measure students' ST learning over time. Specifically, the project will address three questions: (1) What are the standardized ST assessment dimensions that are applicable across STEM fields? (2) Can students' ST learning progressions be measured over time through structural network metrics, functional scenario analyses, and qualitative descriptions defined within semi-quantitative cognitive maps (a) within a specific course and (b) across courses in a curriculum? (3) Are there predictable student learning trajectories or progressions with regard to ST?