Understanding Uncertainty: Exploring Data Using Randomization

In this course we will use computer-intensive methods to explore the randomness inherent in a data set and to develop the scientific logic of statistical inference. We will introduce randomization methods as a basis for framing fundamental concepts of inference: estimates, confidence intervals, and hypothesis tests. The capabilities of computers to draw thousands of random samples and to simulate experiments will replace theoretical approximations grounded in mathematical statistics, especially the normal theory methods like t-tests and chi-squared analyses. Students will use the R programming language to implement the analyses. Much of the course development will proceed through independent and collaborative computer investigations by students using real data sets. No prior experience with statistics and with computer programming is necessary.