In a study of the effects of two anti-anxiety drugs, subjects are asked to rate their anxiety levels under a stressful condition in which they had to perform a complex task. Each of the subjects in the study was given both drugs. Half were given drug A first, and the other half were given drug B first. Their scores under the influence of the two drugs are shown below. Normal distributions CANNOT be assumed. Use a two-tailed test at the .05 level of significance. (8 points)

Drug A

Drug B

15

20

13

17

11

10

12

18

15

13

14

11

11

20

15

22

11

22

16

24

If the distributions were normal what other statistical tests could we use to test for a significant difference between the two groups? How would you expect them to compare with the test you just used and why?