Dear Ann
You have two different way doing this :
- First : each strate as different sample, in that case you must know p1, p2
and alpha for each strate. This is usefull when you want to do complete
analysis in each strate.
- Second : determine the size for the overall population, and then
distribute according to the proportion of each strate in the population. For
instance if you have 100 as the sample size calculated, and 3 strates with
theire respective % : strate1 (30%) , strate2(20%) , strate3(50%)
In your sample you will have : Strate1 : 30%*100=30; strate2: 20%*100=20;
strate3: 50%*100=50.
The second way is usefull when you want to have good estimation of the
overall population.
I hope this will be helpfull for you
Regards
AKC
----- Original Message -----
From: ann e fitzmaurice <[log in to unmask]>
To: <[log in to unmask]>
Sent: Thursday, April 25, 2002 6:27 PM
Subject: query:sample size query
> hi
>
> i would like to compute the sample size requirement for a proposed
> study
>
> i have p1, p2, alpha and power
>
> there may be a complication that the sample may require to be
> stratified by age, how can i build this step into the sample size
> requirement - i have calculated the sample size based on the
> proportion that i have
>
> regards
>
> ann
>
> ----------------------
> ann e fitzmaurice
> medical statistician
> dugald baird centre
> department of obs and gynae
> university of aberdeen
> [log in to unmask]
> tel 01224 553876 (direct)
> fax 01224 404925
|