Fixed n = 8: false positives 5.0%; power at d = 1 46.4%; mean mice per group at d = 1 8.0
Fixed n = 16: false positives 5.1%; power at d = 1 78.1%; mean mice per group at d = 1 16.0
n = 8; if 0.05 <= p < 0.10, add 4 and test again: false positives 6.1%; power at d = 1 56.6%; mean mice per group at d = 1 8.6
The same, second test with cohort as a block: false positives 6.1%; power at d = 1 56.5%; mean mice per group at d = 1 8.6
n = 8; while p >= 0.05 add 4, up to 20: false positives 11.7%; power at d = 1 89.9%; mean mice per group at d = 1 12.1
Planned two-stage design (8, then 16), Pocock p < 0.0294 at each look: false positives 5.1%; power at d = 1 72.7%; mean mice per group at d = 1 13.1