turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

Find a Community

- Home
- /
- SAS Programming
- /
- SAS Procedures
- /
- Can PROC POWER calculate minimum detectable effect...

Topic Options

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

05-20-2015 12:31 PM

Hi,

I am doing a secondary data analysis. I am trying to use PROC POWER to calculate detectable effect size based on extant data. But it seems this SAS procedure can only compute sample size or power rather than detectable effect size. Is that correct?

In addition, SAS can only compute sample size for survival analysis with two exposure groups. Could anyone suggest how to calculate sample size, power or detectable effect size in a survival analysis in which there are 3 exposure groups?

Thanks,

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

07-14-2015 05:55 AM - edited 02-08-2017 03:47 AM

Hello!

As I know 'detectable effect size' need to be based on some historical data or data from previous study.

ICH GCP E9, for example, said on page 17: "The method ...should be given in the protocol, together with the estimates of ... difference to be detected. The basis of these estimates should also be given... The treatment difference to be detected may be based on a judgement concerning the minimal effect which has clinical relevance in the management of patients or on a judgement concerning the anticipated effect of the new treatment, where this is larger." You can show the significance on any data it is the question of the sample size.

About survival analysis for 3 groups I don't know the direct method in SAS, but I use alpha level with a Bonferroni correction (0.05/3 in your way) for solving this kind of problem.

Best regards

Andrey