A Machine Learning Student Behavior Model to Identify Struggling Students in Introductory Computer Science Courses

该项目由罗格斯大学新玩法管理，通过研究人员参与密集的专业成长经验和在SER研究的设计和实施中应用他们的新知识，建立STEM教育研究（SER）的能力。调查员将通过参与重点课程，专业会议和SER专家在项目生命周期中的指导来提高混合方法研究的熟练程度。在专业发展的同时，研究人员将设计和测试一个机器学习工具，以评估学生在计算机科学课程中的真实的时间进度。一旦项目目标实现，开发的工具具有提高学生在计算机科学，在美国劳动力的关键需求领域的成果的潜力。该工具将适用于跨学科使用，以支持学生STEM成果的整体，而研究人员将准备进一步推进有效的STEM教育实践的知识基础。项目研究人员的专业成长活动，通过差距评估定制，将侧重于定性研究方法，学习理论和学习分析。通过学术研究，包括定期与SER专家举行形成会议，并参与重点课程和专业经验，研究人员将建立专业知识，以开发采用行为模型的机器学习工具，预测学生在计算机科学课程中的成功。检测学生在途中的挣扎将促进及时使用干预策略，以提高人口群体的成功，从而提高计算机科学等重要STEM学科的代表性。该项目得到了EHR核心研究（ECR）计划的ECR STEM教育研究竞争能力建设的支持。ECR资助的基础STEM教育研究项目，重点是STEM学习和学习环境，扩大STEM领域的参与，以及STEM专业劳动力的发展。该奖项反映了NSF的法定使命，并已被认为是值得通过使用基金会的智力价值和更广泛的影响审查标准进行评估的支持。

This project, administered by Rutgers University New Brunswick, builds capacity in STEM Education Research (SER) through the investigator’s participation in intensive professional growth experiences and application of their new knowledge in the design and implementation of a SER study. The investigator will build proficiency in mixed methods research by engaging in focused coursework, professional conferences, and mentoring by SER experts over the life of the project. In tandem with professional development efforts, the investigator will design and test a machine learning tool tailored to assess student progress in real time in computer science coursework. Once the project goals are realized, the developed tool holds potential for improving student outcomes in computer science, an area of critical need in the United States workforce. The tool will be adaptable for use across disciplines to support student STEM outcomes overall, while the researcher will be prepared to further advance the knowledge base in effective STEM education practices.The project investigator’s professional growth activities, customized via a gap assessment, will focus on qualitative research methods, learning theory, and learning analytics. Through scholarly study that includes regular formative meetings with experts in SER and engagement in focused coursework and professional experiences, the investigator will build expertise to enable the development of a machine learning tool employing behavioral models that predict student success in computer science courses. Detection of student struggles enroute will promote the timely use of intervention strategies to enhance success across demographic groups, thereby improving representation in vital STEM disciplines such as the computer sciences. This project is supported by the ECR Building Capacity in STEM Education Research competition of the EHR Core Research (ECR) program. ECR funds fundamental STEM education research projects that focus on STEM learning and learning environments, broadening participation in STEM fields, and STEM professional workforce development.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.