From: Multiple imputation by predictive mean matching in cluster-randomized trials
ICC | Cluster size | Clusters | NORM | NORM | NORM |
---|---|---|---|---|---|
ρ | m | k | FE | RE | IGN |
0.03 | 4 | 100 | 95.3 | 96.4 | 96.1 |
8 | 10 | 97.7 | 97.7 | 97.1 | |
20 | 95.8 | 97.0 | 97.0 | ||
40 | 96.5 | 96.2 | 96.4 | ||
40 | 10 | 98.7 | 98.4 | 98.4 | |
20 | 96.4 | 96.5 | 96.8 | ||
40 | 95.7 | 95.8 | 95.9 | ||
400 | 4 | 100 | 100 | 100 | |
0.08 | 4 | 100 | 95.2 | 96.5 | 96.2 |
8 | 10 | 97.6 | 97.6 | 96.9 | |
20 | 95.9 | 96.9 | 96.9 | ||
40 | 96.2 | 95.9 | 96.5 | ||
40 | 10 | 98.6 | 98.3 | 98.4 | |
20 | 96.4 | 96.6 | 96.7 | ||
40 | 95.8 | 95.6 | 96.1 | ||
400 | 4 | 100 | 100 | 100 | |
Bolded values indicate coverage rates below 95%. | |||||
Italicized values indicate coverage rates above 95%. | |||||
Missing data were generated completely at random (MCAR) with a response rate of 60%, and data were imputed using an imputation model that correctly specified the covariate effects. |