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
- /
- Generalization Friedman's test in SAS (PROC freq)

Topic Options

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

Highlighted
# Generalization Friedman's test in SAS (PROC freq)

[ Edited ]
Options

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

01-12-2018 03:57 PM - edited 01-12-2018 03:57 PM

Dear all,

To execute a Friedman's test in SAS, one can use proc freq as explained in the following link:

https://support.sas.com/documentation/cdl/en/statug/63033/HTML/default/viewer.htm#statug_freq_sect033.htm

In addition it is stated that:

*‘If there are multiple subjects per treatment in each block, the ANOVA CMH statistic is a generalization of Friedman’s test.’*

**I am really wondering which generalization is referred to here?** Since the procedure seems to perform fine in a setting with non-equal repeats per treatment in different blocks the only test I am familiar that could fit this framework is the Benard and Van Elteren's test (1953), see e.g. (http://www.sciencedirect.com/science/article/pii/0010480977900040).

Kind regards,

Richard Post MSc

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

Posted in reply to RichardPost

01-13-2018 05:24 AM

Maybe that means non-parameter ANOVA .

proc freq;

table a*b/cmh2 cmb score=row ;

run;

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

Posted in reply to RichardPost

01-19-2018 09:45 AM

The Cochran-Mantel-Haenszel method is a broad, general method for testing in stratified two-way tables. The method includes such things as McNemar's test, Wilcoxon rank sum test, and Cochran's Q, as well as Friedman's test. For more on the method, see the Agresti reference in the References section of the PROC FREQ documentation