Table 5.
Clinician-rated severity group pairwise comparisons (RQ2).
| Severity groups compared | Human WRR |
Machine WRR |
||
|---|---|---|---|---|
| t | p | t | p | |
| Normal & mild | 1.00 | .321 | 1.24 | .220 |
| Normal & moderate | 4.15 | < .001 | 6.48 | < .001 |
| Normal & severe | 8.54 | < .001 | 13.05 | < .001 |
| Normal & profound | 16.89 | < .001 | 15.64 | < .001 |
| Mild & moderate | 2.71 | .009 | 4.50 | < .001 |
| Mild & severe | 6.28 | < .001 | 9.85 | < .001 |
| Mild & profound | 13.41 | < .001 | 12.14 | < .001 |
| Moderate & severe | 3.42 | .001 | 5.09 | < .001 |
| Moderate & profound | 10.59 | < .001 | 7.45 | < .001 |
| Severe & profound | 7.70 | < .001 | 2.61 | .011 |
Note. t statistic (t) and p value (p) for each t test between pairs of clinician-rated severity groups. One set of t tests compared Human word recognition rate (WRR; left), and another compared Machine WRR (right) between each pair of clinician-rated severity groups. All comparisons were evaluated at the α = .005 (.05/10) level, df = 67.