Lesson 11: Investigating the Stroop effect: repeated measures and the paired t-test
Analyse a Stroop experiment with a neutral condition
Skillst-tests and significanceMean, SD and error barsDesigning an investigation
Worked example
Six students named the ink colours of 20 rows of XXXX (neutral) and of 20 incongruent colour words, in counterbalanced order. Times in seconds:
| Student | 1 | 2 | 3 | 4 | 5 | 6 |
|---|---|---|---|---|---|---|
| Neutral | 12.4 | 10.9 | 14.2 | 11.8 | 13.5 | 12.0 |
| Incongruent | 17.9 | 15.2 | 21.0 | 16.2 | 19.4 | 16.6 |
(a) Calculate each student's difference (incongruent − neutral) and the mean difference. (b) Identify the appropriate t-test and its degrees of freedom. (c) The test gives . State the conclusion.
Step 1: The differences
s. Every student was slower on the incongruent card.
Step 2: The mean difference
s.
Step 3: Choose the test
Each student did both conditions (repeated measures), so the data are paired: a paired t-test, .
Step 4: Decide and conclude
: reject H₀. Students took significantly longer to name the ink colours of incongruent words (mean 17.72 s) than of neutral XXXX rows (mean 12.47 s), a mean difference of 5.25 s: a Stroop effect.
The neutral rows have no word to read, so the extra time on the incongruent card is interference from reading the word.
Answer
(a) 5.5, 4.3, 6.8, 4.4, 5.9, 4.6 s; mean difference 5.25 s. (b) A paired t-test with df = 5. (c) : the incongruent words took significantly longer to name than the neutral rows; reject H₀.
Try a similar one
A fresh question of the same kind, checked as you answer, with hints and a full solution.
Method
Subtract each participant's two scores in the same order.
Paired data: paired t-test, .
Decision (compare with 0.05), then a conclusion in words with the means.
Watch out for
Using , the unpaired value.
Subtracting in different orders for different students.