Table 5.
Comparison of design parameters and resulting expected sample sizes between (1) the first design in Table 1 in Whitehead et al. (2009), (2) the δ-minimax design with n1 constrained to be 45, and (3) the δ-minimax design with n1 unconstrained
| Whitehead's | δ-minimax n1 = 45 | δ-minimax n1 unconstrained | |
|---|---|---|---|
| n1 per arm | 45 | 45 | 54 |
| f | 0 | 0.850 | 1.110 |
| e1 | 2.730 | 2.367 | 2.303 |
| n2 per arm | 39 | 52 | 42 |
| e2 | 1.977 | 2.023 | 2.018 |
(N | δ = 0) |
128.75 | 109.75 | 118.45 |
(N | δ = 0.5) |
152.71 | 139.73 | 140.38 |
(N | δ = 1) |
146.79 | 142.28 | 140.71 |
(N | δ = 0)