Data science techniques have revolutionized many academic fields and led to terrific gains in the commercial sector. They have to date been underutilized in solving critical problems in the US educational system, particularly in understanding Science, Technology, Engineering and Mathematics (STEM) learning and learning environments, broadening participation in STEM, and increasing retention for students traditionally underserved in STEM. The goals of the Directorate for Education and Human Resources (EHR), through the EHR Core Research program, for the Critical Techniques and Technologies for Advancing Foundations and Applications of Big Data Science & Engineering (BIGDATA) program are to advance fundamental research aimed at understanding and solving these critical problems, and to catalyze the use of data science in Education Research. Deep learning is a relatively novel method in machine learning (ML) that has shown great improvements over prior ML approaches in successful classification with very little a priori definitions applied to the data. For example, in image classification, these techniques have a high rate of success at distinguishing species and particular animals. This exploratory proposal will investigate the possibilities of using this approach with educational data from Massive Open Online Courses (MOOCs) and Learning Management Systems (LMSs). Research has shown that it is often difficult to use traditional study designs with MOOC data as enrollment is open and many different types of people enroll in MOOCs. This factor has made it difficult to make these MOOCs adaptive to individual learners, an optimal approach to improving learning. The Principal Investigators will group people in MOOCs who are similar to each other to understand the different types of people taking the MOOCs so that learning activities can be tailored to them.
The Principal Investigators will examine a variety of data, from the micro-level of backend data on timing to complete an exercise, or pauses between activities, to macro-level data such as course history and grades. They will use deep learning techniques to identify groups of learners with similar characteristics, the first step in making a learning environment more adaptive. The proposed research is ambitious and risky. Deep learning techniques show great promise and the Principal Investigators demonstrate that they could be extremely helpful in solving key educational challenges. The Principal Investigators have the technical and learning science expertise to carry out this ambitious endeavor. As more and larger educational datasets are developed, the field needs to expand its methodologies to learn from them. This proposal is unique in its potential for catalyzing this effort.
This award is supported by the EHR Core Research (ECR) program. The ECR program emphasizes fundamental STEM education research that generates foundational knowledge in the field. Investments are made in critical areas that are essential, broad and enduring: STEM learning and STEM learning environments, broadening participation in STEM, and STEM workforce development.
