I am actually working on an exercise for my students dealing with multiple comparisons. However, I was having trouble trying to simulate some data in R to demonstrate the problem with multiple comparisons. I was hoping someone might have some R code or at least give some guidance on how to generate a dataset in R to demonstrate the multiple comparisons problem. So first I would like to generate some data (rnorm(), runif, etc.). Then I would like to show how if I pairwise compare the parameters for 2 levels of a multiple level predictor, I am more likely to get a significant result than the standard alpha level of 0.5, or 95% confidence.