Ateacher wants to see if a new unit on fractions is students learn. she has five randomly selected students take a pre-test and a post test on the material. the scores are out of 20. suppose that you are about to compute a confidence interval for hd, how do you check for normality? find the difference between the two scores for each subject. make a plot of the differences and check for outliers. make a plot of the pre and post test scores and see if there are any outliers you don't have to check for normality for this type of data. you don't have to check for normality since the values are large enough.