Again, we focus on statistical inference for randomized experiments. We finish Fisher’s approach as we move from testing a hypothesis about no effects to hypotheses about some effects (and thereby produce confidence intervals). And we think about least squares and average treatment effects (estimating causal effects instead of testing hypotheses about them) from the perspective of a randomized study (both simply randomized and pre-stratified or blocked). Here is the pdf version of the classwork for today with comments . Here is the R source code of the classwork for today .