Continuing the last exercise, plot the difference scores using the R function akerd and note that the distribution is skewed to the right, suggesting that the confidence interval for the mean might be inaccurate.

Using the R function trimcibt, verify that the .95 confidence interval for the 20% of the difference scores is (0.127, 0.543). Then verify that when using the mean instead (setting the argument tr=0), the .95 confidence interval is (0.0097, 0.7969). Why is this second confidence so much longer than the first? Hint: Look at a boxplot of the difference scores.