Suppose you had to identify the color of ink for a series of printed words that spelled out a color that did not match the ink color? For example, what color ink is used in the word RED? This might take longer than identifying the color when the ink and printed word matched (RED). This difference results from interference and is called the Stroop Effect, in recognition of a famous study conducted by research psychologist J. R. Stroop (1897-1973).
Same Different
32 ...........66
12 ...........31
17 ...........40
16 ...........25
25 ...........36
18 ...........15
18 ...........39
24 ...........35
20 ...........32
24 ...........30
The data in the table were collected to see whether interference was significant. Each of the 10 people studied identified colors of ink in two situations, and the time (in seconds) was recorded. There were 36 words in each trial. In column 1 (Same), the color of the ink was consistent with the meaning of the spelled-out word. In column 2 (Different) the color of the ink was different from the meaning of the spelled-out word. Each person was randomly assigned to read the "same color" or the "different color" words first. Assume that all the conditions required for t-tests are satisfied.
a. Compare the sample means. Does the result fit the theory given?
b. Use a two-sample t-test; report the t-statistic and p-value, and report whether you can reject the null hypothesis of no difference in times at the 0.05 level.
c. Use a paired t-test; report the t-statistic and p-value, and report whether you can reject the null hypothesis of no difference in times at the 0.05 level.
d. Which is the appropriate test for this data set?
e. Why is it appropriate to randomly assign the order so that some of the people read the "same color" word first and some read the "different color" word first?