Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Paired proportions- power calculation

From   "b. water" <>
Subject   Re: st: Paired proportions- power calculation
Date   Tue, 26 Aug 2003 12:44:46 +0000


I do not know how to do it in Stata. Maybe it can be done (with or without ease), maybe not. In the past when I have to do power calculations on paired proportions - I used a software called power and precision. The manual that comes with the trial version (the 7 days trial version allows for 10 calculations) also included the formulas, which, I supposed, if you are proficient in Stata you can use adapt it to Stata. Here's the link to the trial version of the software:

Apologies to other Statalisters for listing or suggesting non-Stata related software. I adopt a pragmatic approach to this issue in my working life. Although I am learning Stata, in the meantime, if I came across a problem that I know can be better handled by other software (perhaps due to my shortcoming in Stata) then I'll used that software in the interim - to save time and effort. Till such time I can use Stata for that problem. Imho. Hope this helps.


From: "Andrew McIntosh" <>
To: <>
Subject: st: Paired proportions- power calculation
Date: Mon, 25 Aug 2003 21:13:48 +0100

I (will) have a sample of 30 pairs of people and I want to calculate the

power to detect a difference of 5, 10, or 15 events between the two
(paired ) groups, assuming the control group have an event rate of about

Unforunately, the 'sampsi' command doesn't seem to extend to matched
pairs and I wondered if there was a method of completing this
analysis using another STATA command? Thank you for any help...

Andrew McIntosh, Edinburgh University

*   For searches and help try:
MSN 8: Get 6 months for $9.95/month

*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index