S022: Introduction to Statistical Computing and Data Science in Education

This course focuses on applying modern data science and machine learning tools to real-world datasets in education. We first teach tools for exploring new datasets in order to identify patterns, make predictions from flexible models, and visualize data in ways that communicate complex associations. We also expand on the core conceptual building blocks taught in S-40 to provide more flexible approaches to estimation and inference, with a particular focus on the bootstrap. Throughout, we will learn statistical computing in R, an increasingly important skill in the modern, data-driven era. By the end of the course, students will be able to independently analyze data of various types, carrying a project from getting the data ready for analysis to creating technical reports of one’s findings. Topics covered include classification and regression trees, random forests, regularized regression, cross-validation, data wrangling, model selection, bootstrapping, and data visualization. While we assume foundational statistical knowledge, we do not assume any initial familiarity with statistical computing or the R language. Students interested in getting started with learning R before the course starts should contact the instructor.