The following data show the pattern of results that was obtained in a study by Liguori and Robinson (2001) examining how different levels of alcohol and caffeine consumption influenced response time in a simulated driving test. The means show the average response time in milliseconds for different combinations of alcohol and caffeine. For these data:

a. Is there a main effect for alcohol? Yes.

b. Is there an interaction? Yes.

c. Does caffeine improve response time (produce faster times) for people who have consumed alcohol? Yes.

d. Does caffeine eliminate the effect of alcohol on response time? No.