Table 4. Small ruminants Bayesian latent class models results from two-test and three-test models showing parameter medians (95% CrI) for POC-CCA accuracy.
| Site | Variable | Two-Test Independence | Two-Test Dependence |
Three-Tests |
|---|---|---|---|---|
| B | Sensitivity (%) | 49 (29, 87) | 42 (7, 85) | 41 (25, 66) |
| Specificity (%) | 91 (73, 99) | 85 (58, 99) | 91 (75, 99) | |
| Se cov | n.a. | 0 (-0.1, 0.1) | 0.09 (0, 0.16) | |
| Sp cov | n.a. | 0.03 (-0.03, 0.17) | 0.02 (-0.03, 0.12) | |
| DIC | 55.92 | 55.94 | 107.55 | |
| R | Sensitivity (%) | 12 (1, 37) | 13 (1, 40) | 26 (8, 68) |
| Specificity (%) | 88 (65, 99) | 89 (65, 99) | 93 (74, 100) | |
| Se cov | n.a. | -0.02 (-0.13, 0.03) | 0.12 (-0.01, 0.21) | |
| Sp cov | n.a. | 0.02 (-0.03, 0.11) | 0.01 (-0.05, 0.1) | |
| DIC | 46.53 | 55.92 | 92.92 | |
| B & R | Sensitivity (%) | n.a. | 30 (3, 69) | 34 (19, 65) |
| Specificity (%) | 91 (77, 99) | 86 (61, 99) | 92 (80, 99) | |
| Se cov | n.a. | -0.01 (-0.1, 0.07) | 0.12 (0.04, 0.19) | |
| Sp cov | n.a. | 0.03 (-0.02, 0.16) | 0.02 (-0.01, 0.09) | |
| DIC | 54.01 | 54.51 | 106.18 | |
| Pr(Se B > Se R) | 0.99 | 0.91 | 0.80 | |
| Pr(Sp B > Sp R) | 0.60 | 0.41 | 0.45 |
Se = sensitivity; Sp = specificity; cov = covariance; DIC = Deviance information criterion; B = Barkedji; R = Richard Toll; n.a. = not applicable; Pr = probability