"Introducing Inference with Bootstrapping and Randomization"
with Kari Lock Morgan, Duke University
(Tip: click the fullscreen control)
Having trouble viewing? Try: Download as .mp4 (h.264/AAC MPEG-4)
While formulas and distributional tables were once necessary for teaching confidence intervals and hypothesis tests, computers have made simulation methods an attractive alternative. We choose to introduce inference using simulation methods because they require less background knowledge, the same procedure can be applied to any parameter of interest, and most importantly, the methods for generating intervals and p-values are intrinsically connected to the very concepts we most want students to understand. We'll discuss how and why we now use simulation methods to introduce inference, how they can be integrated into the existing curriculum, and outcomes from teaching with this approach. We'll also demonstrate new user-friendly applets for teaching and using simulation methods.
Kari Lock Morgan