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
- /
- Analytics
- /
- Stat Procs
- /
- Somers' D

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

10-21-2007 12:45 PM

This question is about SAS/STAT and I think this forum might be the best fit for my question.

In categorical data analysis, SAS generates Somers' D to test the association strength between categorical variables.

In SAS, both Proc Freq and Proc Logistic generate Somers' D. My question is what is the difference between the Somers' D from Proc Freq and Proc Logistic?

Most of time, Somers' D from Proc Logistic (when only one factor is applied) is quite closed to the Somers' D from Proc Freq. However, they could also be quite different. Could someone tell me what make values from both Procedures so different?

Thanks in advance.

Ben

In categorical data analysis, SAS generates Somers' D to test the association strength between categorical variables.

In SAS, both Proc Freq and Proc Logistic generate Somers' D. My question is what is the difference between the Somers' D from Proc Freq and Proc Logistic?

Most of time, Somers' D from Proc Logistic (when only one factor is applied) is quite closed to the Somers' D from Proc Freq. However, they could also be quite different. Could someone tell me what make values from both Procedures so different?

Thanks in advance.

Ben

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

10-28-2007 09:35 PM

The method of calculation may differ quite markedly for each procedure to allow for the functionality provided by the analysis.

The calculation method for each procedure is explored in great detail, and will make profound sense to someone who understands statistical techniques and the limitations of particular interpretive methods.

If you don't have SAS/Help installed, look up the product documentation on the SAS website, navigate through the product and look at the background material supplied for the appropriate procedure.

Kind regards

David

The calculation method for each procedure is explored in great detail, and will make profound sense to someone who understands statistical techniques and the limitations of particular interpretive methods.

If you don't have SAS/Help installed, look up the product documentation on the SAS website, navigate through the product

Kind regards

David