Combining Machine Learning Explanation Methods with Expectancy-Value Theory to Identify Tailored Interventions for Engineering Student Persistence

该项目旨在通过应用机器学习(ML)方法来提高工科学生的毅力，以服务于国家利益。对多样化、合格的工程劳动力的需求日益增长，这就要求提高本科课程的持久率，特别是对代表性不足和小规模的学生。这个参与式学生学习2级项目旨在应用成熟的预测性ML方法和尖端的ML解释方法来识别和预测哪些学生可能离开以及为什么离开。该项目团队计划确定量身定做的干预措施，以满足个别学生的需求。该项目将测试各种ML方法的可行性和可变性，并将该方法推广到其他大学。拟议的项目有可能产生知识，并简化确定工程持久性个性化干预措施的方法。该项目以期望值理论为基础。拟议的方法将帮助工程教育研究人员在第一年开始干预，帮助处于危险中的学生坚持下去。该项目还有可能对计算机科学界产生广泛的影响，这些社区需要新兴工具在现实世界中的应用。项目成果将在美国工程教育学会年会和教育前沿年会上公布。此外，每个阶段的所有蟒蛇代码都将通过开放科学基金会提供。这些代码将附有对新数据使用拟议方法的详细说明。提出的提高工程持久性的方法可能会在当地、全国或全球范围内产生广泛的影响。NSF IUSE：EDU计划支持研究和开发项目，以提高所有学生的STEM教育的有效性。通过参与的学生学习路径，该计划支持有前景的实践和工具的创建、探索和实施。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

This project aims to serve the national interest by applying machine learning (ML) methods to improve engineering student persistence. The increasing demand for a diverse, qualified engineering workforce requires an improvement in undergraduate program persistence rates, especially for underrepresented and minoritized students. This Engaged Student Learning Level 2 project intends to apply well-established predictive ML methods and cutting-edge ML explanation methods to identify and predict which students might leave and why. The project team plans to identify tailored interventions to meet individual student needs. This project will test the viability and variability of various ML methods and generalize the methodology for other universities. The proposed project has the potential to generate knowledge and streamline a methodology for the identification of individualized interventions for engineering persistence. The project is grounded in expectancy-value theory. The proposed methodology will help engineering education researchers to begin intervening in the first year to help at-risk students persist. This project also has the potential to have broad impacts on computer science communities which require real-world applications of emerging tools. Project findings will be presented at the American Society of Engineering Education annual conference and Frontiers in Education annual conference. In addition, all python code for each stage will be made available through the Open Science Foundation. The codes will be accompanied with detailed instructions for using the proposed methodology with new data. The proposed methods to improve engineering persistence potentially have broad impacts in a local, nation-wide, or global scale. The NSF IUSE: EDU Program supports research and development projects to improve the effectiveness of STEM education for all students. Through the Engaged Student Learning track, the program supports the creation, exploration, and implementation of promising practices and tools.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.