Example Scenario
Suppose we want to test whether a new study technique improves test
scores. We have test scores of 5 students before and after using the
technique.
Student Before (X) After (Y)
1 78 85
2 82 88
3 75 80
4 90 92
5 88 85
Step 1: Calculate Differences
Calculate the difference (D_i = X_i - Y_i) for each student.
Student Before (X) After (Y) Difference (D = X - Y)
1 78 85 -7
2 82 88 -6
3 75 80 -5
4 90 92 -2
5 88 85 3
Step 2: Rank the Absolute Differences
Now, we take the absolute values of the differences and rank them.
Difference
Student Absolute Difference ( D ) Rank
(D)
1 -7 7 5
2 -6 6 4
3 -5 5 3
4 -2 2 2
5 3 3 1
Step 3: Assign Signs to Ranks
Now, we assign the sign of the original difference to the ranks.
Student Difference (D) Rank Signed Rank
1 -7 5 -5
2 -6 4 -4
3 -5 3 -3
4 -2 2 -2
5 3 1 1
Step 4: Calculate the Test Statistic
Now, we calculate the sum of the positive ranks and the sum of the negative
ranks.
Sum of positive ranks ((W^+)): (1) (from Student 5)
Sum of negative ranks ((W^-)): (-5 - 4 - 3 - 2 = -14)
The test statistic (W) is the smaller of the absolute values of these sums: [W
= \min (1, 14) = 1]
Step 5: Determine the Critical Value or p-value
For a sample size of 5 (number of non-zero differences), we can refer to a
Wilcoxon signed-rank table or use statistical software to find the critical value
for (W) at a significance level (e.g., (\alpha = 0.05)).
For (n = 5), the critical value for a two-tailed test is typically around 0 or 1.
Step 6: Make a Decision
If (W) is less than or equal to the critical value, we reject the null
hypothesis.
In this case, (W = 1) is equal to the critical value, so we reject the null
hypothesis.
Step 7: Report on the Results
We conclude that there is a statistically significant difference in test scores
before and after using the new study technique.