Question

A study of the impact of caffeine consumption on reaction time was designed to correct for the impact of subjects’ prior sleep deprivation by dividing the $24$subjects into $12$pairs on the basis of the average hours of sleep they had had for the previous 5 nights. That is, the two with the highest average sleep were a pair, then the two with the next highest average sleep, and so on. One randomly assigned member of each pair drank $2$cups of caffeinated coffee, and the other drank $2$cups of decaf. Each subject’s performance on a Page Number: $690$standard reaction-time test was recorded. Which of the following is the correct check of the “Normal/Large Sample” condition for this significance test?

I. Confirm graphically that the scores of the caffeine drinkers could have come from a Normal distribution.

II. Confirm graphically that the scores of the decaf drinkers could have come from a Normal distribution.

III. Confirm graphically that the differences in scores within each pair of subjects could have come from a Normal distribution.

a. I only

b. II only

c. III only

d. I and II only

e. I, I, and III

Short answer

The correct option is c.

c.) III only - Confirm graphically that the differences in scores within each pair of subjects could have come from a Normal distribution.

Step-by-step solution

Step 1 -  Given information:

We have been given that:

No. of subjects = $24$

No. of pairs =$12$

Each pair had one member drink two cups of caffeinated coffee and the other two cups of decaf.

Step 2 - Explanation:

The Large Enough Sample Condition evaluates whether your sample size is adequate for the population.

A common rule of thumb for the Large Enough Sample Condition is $n\ge 30$, where n is your sample size.