Table 1.
Frequency of misclassifications between the AI categories derived from the computer graphics tablet scores and examiners’ categories in each session
| n (%) correct observations | n (%) misclassifications, real AI higher than examiner’s score | n (%) misclassifications, real AI lower than examiner’s score | |
|---|---|---|---|
| Examiner 1 |
|
|
|
| Session 1 |
55 (92) |
4 (7) |
1 (2) |
| Session 2 |
53 (88) |
6 (10) |
1 (2) |
| Examiner 2 |
|
|
|
| Session 1 |
49 (82) |
10 (17) |
1 (2) |
| Session 2 | 51 (85) | 8 (13) | 1 (2) |