⏳ Loading example…
🔢
Non-parametric
Wilcoxon Signed-Rank Test — Paired Non-parametric
Non-parametric alternative to the paired t-test. Before/after coping scores for 9 employees after a mindfulness program, using the exact signed-rank distribution (n<50, no ties).
📊 W = 2, n = 9, p = .012 (exact)
Step-by-step solution
🔢 30 · Wilcoxon Signed-Rank Test — Paired Non-parametric
Before vs. After an 8-Week Mindfulness Program — Coping Scores
Non-parametric
Research Question:
Did coping scores (0–100 scale) change for the same 9 employees after an 8-week mindfulness program? The scores are skewed, so a paired t-test's normality assumption is doubtful — the Wilcoxon signed-rank test is the non-parametric alternative for paired data.
| Employee | Before | After | Diff (After−Before) |
|---|---|---|---|
| 1 | 60 | 66 | +6 |
| 2 | 58 | 62 | +4 |
| 3 | 65 | 74 | +9 |
| 4 | 72 | 70 | −2 |
| 5 | 55 | 62 | +7 |
| 6 | 63 | 66 | +3 |
| 7 | 59 | 64 | +5 |
| 8 | 68 | 76 | +8 |
| 9 | 61 | 62 | +1 |
Sum of ranks for positive and negative differences
-
1Rank the absolute differences |d| from smallest to largest (no ties here):
-
2Sum the ranks separately for positive and negative differences: Only employee #4 got worse (diff = −2, rank 2); every other employee improved.
-
3Exact test (n=9 ≤ 49, no ties, no zero differences): MindStat and R's wilcox.test() both use the exact distribution of W under H₀ whenever n < 50 with no ties — the normal approximation is only a fallback for larger or tied samples.
-
4Decision rule: p = .012 < .05 → reject H₀ → coping scores changed significantly after the program.
2
W
9
n
.012
p (exact)
🔴 Significant — W = 2, n = 9, p = .012 (exact, two-tailed). Coping scores increased after the program.
APA-7
A Wilcoxon signed-rank test indicated that coping scores were significantly higher after the mindfulness program than before, W = 2, p = .012 (exact, two-tailed).