With Andrew Bray (Reed College)
A first course in statistics often exhibits a tension between the conceptual understanding of inference and the practical understanding of how to carry out inference on a real data set. The infer package in R was created to address this tension with an expressive and intuitive computational framework for statistical inference. In this breakout session we will discuss the design principles of the package, which unite resampling and approximation-based approaches and are motivated by the emerging ecosystem for data science called the “tidyverse”. Session attendees will have the opportunity to experiment with the package to construct intervals and conduct tests based on data from the General Social Survey.