Log in
RegisterLost sequence [Design Issues]
Dear John,
Nothing serious happens even in that highly unlikely case
.
The study stays evaluable.
The only thing you loose is that the design is no longer a design with minimum variance. I.e. the confidence intervals a somewhat broader compared to a balanced design with the same number of subjects (if this exists).
Just to cite our Ol' pirate and captain ElMaestro
:
"Imbalance, as I see it, no doubt plays a role - sometimes even an obscure one ..."
❝ But what if I end up losing all my subjects from 1 of the 6 sequence? I know it's highly unlikely...
Nothing serious happens even in that highly unlikely case
.The study stays evaluable.
The only thing you loose is that the design is no longer a design with minimum variance. I.e. the confidence intervals a somewhat broader compared to a balanced design with the same number of subjects (if this exists).
Just to cite our Ol' pirate and captain ElMaestro
:"Imbalance, as I see it, no doubt plays a role - sometimes even an obscure one ..."
—
Regards,
Detlew
Regards,
Detlew
Complete thread:
- 3 period BE studies, 3 or 6 sequence jag009 2012-06-06 22:18
![[Mix]](https://static.bebac.at/img/mix.png "open in Mix view")
- 3 period BE studies, 3 or 6 sequence Helmut 2012-06-07 13:48
- 3 period BE studies, 3 or 6 sequence drgunasakaran1 2012-06-07 17:48
- 3 period BE studies, 3 or 6 sequence jag009 2012-06-07 20:02
- Lost sequenced_labes 2012-06-08 08:49
- 3 period BE studies, 3 or 6 sequence jag009 2012-06-07 20:02
Ing. Helmut Schütz