The aim of this course is to provide students with the understanding and tools to critically consume and conduct statistical research. The theme is the challenge of drawing reliable causal inference. We will learn: how to use graphical methods to transparently analyze and present data; how to discipline our analyses against multiple-comparisons bias; how to use nonparametric methods to avoid implausible assumptions; how strong research design is essential to causal inference; how Bayesian inference provides the mathematical vocabulary for thinking about scientific inference; how causal graphs allow us to express and analyze causal assumptions, choose control variables, and think about selection bias; how placebo tests allow us to test assumptions; how to build and understand Likelihood and Bayesian models including Logistic and Probit models; how to think about and analyze time-series cross-sectional data. We will review instrumental variables methods and regression-discontinuity designs, though it is assumed that you have already covered these in PLSC 503.