Table 2 The influence of data quality on the association between antibiotics used and clinically restored on day 21 in Example 2: Moderate coliform mastitis. One of the clinical evaluations has been changed and the influence on the significance level (p-value) is given. Table A is the original data, tables B-D are modifications. Figures in boldface and italic are clinical examinations that have been changed.
From: Epidemiological Studies Based on Small Sample Sizes – A Statistician's Point of View
A | Clinically restored | ||
|---|---|---|---|
Yes | No | ||
Antibiotic | Yes | 7 | 0 |
No | 2 | 4 | |
Test for association : p = 0.020. | |||
B | Clinically restored | ||
Yes | No | ||
Antibiotic | Yes | 6 | 1 |
No | 2 | 4 | |
Test for association: p = 0.100 | |||
C | Clinically restored | ||
Yes | No | ||
Antibiotic | Yes | 7 | 0 |
No | 1 | 5 | |
Test for association: p = 0.005 | |||
D | Clinically restored | ||
Yes | No | ||
Antibiotic | Yes | 7 | 0 |
No | 3 | 3 | |
Test for association: p = 0.070 | |||