Nope [Two-Stage / GS Designs]
IANAL, but I think what San-Diego-man is proposing is to calculate BE-outcome (pass/fail) after stage 1 and compare it to BE-outcome after the (unnecessary) second stage for, let's say, 1 gazillion of studies and see if there is a meaningful difference.
Isn't this a special example of the more general "forced BE" theme in one-stage designs?
Isn't this a special example of the more general "forced BE" theme in one-stage designs?
—
Kindest regards, nobody
Kindest regards, nobody
Complete thread:
- Two PK metrics: Inflation of the Type I Error Helmut 2017-11-12 11:57 [Two-Stage / GS Designs]
- Two PK metrics: Inflation of the Type I Error? d_labes 2017-11-12 17:46
- A place to start ElMaestro 2017-11-12 21:43
- A place to start? d_labes 2017-11-13 16:02
- A place to start? Helmut 2017-11-13 16:51
- Scientific gut feeling d_labes 2017-11-15 10:34
- Five minutes gone - power.tsd.2m() arose d_labes 2017-11-15 13:15
- Scientific gut feeling d_labes 2017-11-15 10:34
- A better place to start. ElMaestro 2017-11-14 00:04
- Nope Helmut 2017-11-14 00:21
- Nopenobody 2017-11-14 08:11
- I've meditated hard ElMaestro 2017-11-14 12:41
- Nope Helmut 2017-11-14 00:21
- A place to start? Helmut 2017-11-13 16:51
- A place to start? d_labes 2017-11-13 16:02