16. Non-parametric Tests
16.1 How to Rank Data
Many of the non-parametric tests that we’ll look at require that you rank data. Here we review the conventions for ranking that we need :
- Assign rank from the lowest score to the highest score.
- If there are ties, assign the average rank to all ties.
Example 16.1 :
Subject | Score | Rank |
A | 8 | 4 |
B | 6 | 3 |
C | 10 | 5 |
D | 3 | 2 |
E | 1 | 1 |
▢
Example 16.2 :
Subject | Score | Rank |
A | 8 | 4 |
B | 6 | 2.5 |
C | 10 | 5 |
D | 6 | 2.5 |
E | 1 | 1 |
B and D are tied for 2nd and 3rd place, so they are ranked at the average of 2 and 3, 2.5.
To determine the ranking, it may help to sort the data based on rank :
Subject | Score | Rank |
E | 3 | 1 |
D | 6 | 2.5 |
B | 6 | 2.5 |
A | 8 | 4 |
C | 10 | 5 |
This way ties are easier to see.
▢