Table 2.
Mean squared errors of the OR and CI width for the OR from simulation study of the probabilistic bias analysis and Bayesian approaches to correcting for misclassification.
| Truth: p1=.6 | Truth: p1=.8 | |||
|---|---|---|---|---|
| Bayes | Probabilistic bias analysis | Bayes | Probabilistic bias analysis | |
| Vaguely Informative | ||||
| MSE | 0.14 | 0.40 | 0.02 | 0.07 |
| Widtha | 12.3 | 5.1 | 11.4 | 5.9 |
| Weakly Informative | ||||
| MSE | 0.04 | 0.19 | 0.04 | 0.03 |
| Width a | 14.6 | 6.9 | 10.4 | 5.8 |
| Highly Informative | ||||
| MSE | 0.09 | 0.07 | 0.03 | 0.03 |
| Width a | 7.0 | 4.4 | 1.6 | 1.4 |
a
The difference between the upper and lower bounds of the 95% CI for the OR