Good point! [Power / Sample Size]
good points! First of all I could reproduce your numbers; nice excursion into combinatorics.
library(PowerTOST)
CV <- 0.262
dor <- 0.15
n.des <- sampleN.TOST(CV=CV, details=F, print=F)[["Sample size"]]
n.adj <- round(n.des/(1-dor)/2)*2
do <- n.adj-n.des
n.TR <- seq(n.adj/2-do, n.adj/2, 1)
n.RT <- seq(n.adj/2, n.adj/2-do, -1)
res <- matrix(nrow=length(n.TR), ncol=7, byrow=T, dimnames=list(rep(NULL, 5),
c("n", "TR", "RT", "TR.do", "RT.do", "prob", "power")))
res[, 1] <- n.TR+n.RT
res[, 2] <- n.TR
res[, 3] <- n.RT
res[, 4] <- n.adj/2-n.TR
res[, 5] <- n.adj/2-n.RT
for (j in seq_along(n.TR)) {
res[j, 6] <- sprintf("%.9f",
(choose(n.adj/2, n.adj/2-n.TR[j])*
choose(n.adj/2, n.adj/2-n.RT[j]))
/choose(n.adj, do))
res[j, 7] <- sprintf("%.5f", power.TOST(CV=CV, n=c(n.TR[j], n.RT[j])))
}
Sum.p <- sum(as.numeric(res[, 6]))
res <- data.frame(res)
print(res, row.names=F); cat("Sum p:", Sum.p, "\n")
n TR RT TR.do RT.do prob power
30 12 18 6 0 0.009530792 0.78425
30 13 17 5 1 0.079178886 0.79345
30 14 16 4 2 0.240364474 0.79877
30 15 15 3 3 0.341851697 0.80051
30 16 14 2 4 0.240364474 0.79877
30 17 13 1 5 0.079178886 0.79345
30 18 12 0 6 0.009530792 0.78425
Sum p: 1
❝ So in this example we have 65.8% (100*(1-p_4)) probability that we shoot ourselves in the foot .
Depends on how rigid the bullet is. Full metal jacket are not even the 6/0 (or 0/6)-cases (power 78.4%). The others (with a much higher probability) are soft-balls.
❝ Maybe not so big shot - planned number of subjects finished will be achieved , but sample size wasn't estimated with option of getting unbalanced in mind. Even though there is almost 2/3 probability that it will ends unbalanced (if assumptions are right, and no better GMR than expected, no lower CV estimated from residual mean square than expected, and no lower dropout-rate than expected).
Correct. I would be wary to assume very different dropout-rates in sequences. Theoretically they should occur at random (nTR ~ nRT). If we expect different dropout-rates a priori, IMHO this would also imply that we expect a true (!) sequence effect – which would confound the treatment effect. Opening a can of worms.

❝ "Remember, with great power comes great responsibility."
Power. That which statisticians are always calculating
but never have. Stephen Senn (Statistical Issues in Drug Development, Wiley 2004, p197)
Dif-tor heh smusma 🖖🏼 Довге життя Україна!
![[image]](https://static.bebac.at/pics/Blue_and_yellow_ribbon_UA.png)
Helmut Schütz
![[image]](https://static.bebac.at/img/CC by.png)
The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
Complete thread:
- Adjusting for expected dropout-rate Helmut 2015-06-17 12:33 [Power / Sample Size]
- Percentage calculation d_labes 2015-06-18 09:02
- Adjusting for expected dropout-rate - Nitpicking zizou 2015-08-18 23:34
- Good point!Helmut 2015-08-19 02:20
- Good point! intuitivepharma 2015-08-19 07:47
- Copy paste error d_labes 2015-08-19 10:12
- Copy paste error intuitivepharma 2015-08-19 11:18
- PowerTOST <1.2.7? Helmut 2015-08-19 11:44
- PowerTOST <1.2.7? intuitivepharma 2015-08-26 14:41
- PowerTOST <1.2.7? Helmut 2015-08-19 11:44
- Copy paste error intuitivepharma 2015-08-19 11:18
- Copy paste error d_labes 2015-08-19 10:12
- Good point! intuitivepharma 2015-08-19 07:47
- Good point!Helmut 2015-08-19 02:20