Dear Statalister (Joe Labarere)
Since the release of sskdlg I have spotted this bug and fixed it. Here is
what I get with the my corrected edition of sskdlg:
--- Begin -----------------------------------------------
Results for kappa=0.9, p1=0.7, p2=0.7,
d=0.05 [95% Conf. Interval]:
* Sample size = 350
If the sample size is fixed at 350,
given the kappa, p1, p2 and CI stated
above, d = .05.
--- End -------------------------------------------------
This corrected version has been sent to the editor last year but apparently
it has not been made available yet. I'll contact the editor again. In the
mean time, if you wish, I can send you the fixed version of the ado-file in
private. Please contact me on the e-mail below.
Sincerely,
Michael E. Reichenheim
Instituto de Medicina Social/UERJ
[email protected]
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]]On Behalf Of Labarere, Jose
> Sent: quinta-feira, 31 de outubro de 2002 08:59
> To: [email protected]
> Subject: st: problem with sskdlg ado
>
>
> Dear statalisters,
>
> Does anyone know why the absolute precision d (0.05) is different when
> estimating sample size (350) or using the "sample size for d" option of
> sskdlg (d=0.036 wit a preset sample size of 350) ? see below
>
> I'd appreciate any help with this.
>
> Sincerely,
> J Labarere
>
> --- Begin -----------------------------------------------
>
>
> Results for kappa=.90, p1=0.70, p2=0.70,
>
> d=0.05 [95% Conf. Interval]:
>
>
> * Sample size = 350
>
>
>
>
> If the sample size is fixed at 350,
>
> given the kappa, p1, p2 and CI stated
>
> above, d = .036.
>
>
> --- End -------------------------------------------------
>
>
>
> *
> * For searches and help try:
> * http://www.stata.com/support/faqs/res/findit.html
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
>
>
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
