Table 2.
Correspondence between confidence classification methods in the online dataset.
| Classification Method #1 | Measure | Confidence Classification Method #2 | |||
|---|---|---|---|---|---|
| Human Coder 1 | Human Coder 2 | AI Model | Trichotomized | ||
| Human Coder 1 | Correlation (r) | - | |||
| % agreement | - | ||||
| Cohen’s k | - | ||||
| Human Coder 2 | Correlation (r) | 0.695 | - | ||
| % agreement | 72.4% | - | |||
| Cohen’s k | 0.577 | - | |||
| AI Model | Correlation (r) | 0.876 | 0.720 | - | |
| % agreement | 87.9% | 72.8% | - | ||
| Cohen’s k | 0.790 | 0.588 | - | ||
| Trichotomized Numeric Confidence | Correlation (r) | 0.710 | 0.614 | 0.732 | - |
| % agreement | 58.0% | 43.0% | 58.0% | - | |
| Cohen’s k | 0.620 | 0.438 | 0.641 | - | |
Note. N = 1072.