Accelerating STEM Learning Through Large-Scale Data Science

在NSF加速发现计划的支持下，莱斯大学的研究人员将设计和实施教育研究网络（LERN）学习，以研究和利用推动人类学习的复杂因素。 其中心思想是，精心设计的、涉及大量学生和教师的大规模学习研究，加上现代数据科学工具，可以深入了解支撑学生学习的因素。 长期以来，研究人员一直试图确定学术成功和持久性的预测因素，因为这些预测可以为有失败风险的学生提供先发制人的干预。 然而，这些研究通常是针对学生群体而不是个人。 人口的重点限制预测的学生群体，而不是个人。 LERN项目旨在利用快速的技术进步，开发个性化的学习研究方法，将学习科学从基于人口的思维转向基于个人的思维。 通过这种方式，该项目旨在为个性化学习做出贡献，因此可以使用个性化定制的方法来解决每个学生在学习和成就方面的差异。LERN研究的基于个人的方法的特点是强调个体差异如何影响学习成果，以及这些差异如何与学习干预相互作用。 该项目旨在开发：（1）新的数据科学工具，以衡量学生的个体差异及其与真实数字学习环境的互动;（2）大规模的基础设施，以科学地操纵学生的学习体验;（3）一套有针对性的实验和研究，利用大量的学生;（4）将学生个体差异和学习互动数据与社会经济和人口统计数据相融合的研究。 LERN将使用数字学习平台OpenStax，研究德克萨斯州休斯顿大量学生学习中的多因素交互作用。 该项目旨在开发一套数据科学工具，包括自适应评估和数字突出显示和注释，这将允许检查个体差异和学习行为，包括控制点和学术心态。 这些努力旨在将教育研究的进程从以人口为基础转向以个人为基础，从而为个性化学习奠定基础。 这些努力补充了美国国家科学基金会对利用数据革命的关注，利用数据科学方法来理解和改善学生的学习。该奖项反映了NSF的法定使命，并被认为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来支持。

With support from NSF's Accelerating Discovery program, investigators at Rice University will design and implement the Learning in Education Research Network (LERN) to study and harness the complex factors that propel human learning. The central idea is that carefully designed, large-scale learning studies involving large numbers of students and teachers, coupled with modern data science tools, can provide insight into the factors that underpin student learning. Researchers have long sought to determine the predictors of academic success and persistence since such predictions could enable preemptive interventions for students at risk for failure. However, these studies have usually examined populations of students rather than individuals. The population focus limits prediction to groups of students rather than to individuals. The LERN project intends to take advantage of rapid technological advances to develop individually tailored ways of studying learning, shifting the science of learning away from population-based thinking and towards individual-based thinking. In this way, the project intends to contribute to personalized learning, so individually tailored approaches can be used to address each student's differences in learning and achievement.The individual-based approach to be studied in LERN is characterized by an emphasis on how individual differences affect learning outcomes, and how these differences interact with learning interventions. The project intends to develop: (1) new data science tools to measure students' individual differences and their interactions with authentic digital learning environments; (2) a large-scale infrastructure to scientifically manipulate students' learning experiences; (3) a suite of targeted experiments and studies that leverage large numbers of students; and (4) a study fusing student individual differences and learning interaction data with socioeconomic and demographic data. LERN will use OpenStax, a digital learning platform, to study multiple factor interactions in learning across large numbers of students in Houston, Texas. The project intends to develop a set of data science tools, including adaptive assessment and digital highlighting and annotation, that will allow examination of individual differences and learning behaviors including locus of control and academic mindset. These efforts are intended to shift the course of educational research from population-based to individual-based, thus setting the stage for personalized learning. These efforts complement the National Science Foundation's focus on Harnessing the Data Revolution by harnessing data science approaches to understand and improve student learning.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.