https://communities.sas.com/t5/Statistical-Procedures/post-chi-square-test/m-p/61646#M2883 I am doing a chi-square test to see if there is a difference in the membership renewal rates among multiple groups.<BR /> <BR /> The chi-square test is significant. Now, if I want to test which cells are different, how do i do the marascuilo procedure in SAS? I found out from some papers that I can do this with PROC GENMOD and PROC MULTTEST. Any suggestions?<BR /> <BR /> Thanks Message was edited by: datalligence Wed, 05 Aug 2009 13:49:08 GMT datalligence 2009-08-05T13:49:08Z https://communities.sas.com/t5/Statistical-Procedures/post-chi-square-test/m-p/61646#M2883 I am doing a chi-square test to see if there is a difference in the membership renewal rates among multiple groups.<BR /> <BR /> The chi-square test is significant. Now, if I want to test which cells are different, how do i do the marascuilo procedure in SAS? I found out from some papers that I can do this with PROC GENMOD and PROC MULTTEST. Any suggestions?<BR /> <BR /> Thanks Message was edited by: datalligence Wed, 05 Aug 2009 13:49:08 GMT https://communities.sas.com/t5/Statistical-Procedures/post-chi-square-test/m-p/61646#M2883 datalligence 2009-08-05T13:49:08Z https://communities.sas.com/t5/Statistical-Procedures/post-chi-square-test/m-p/61647#M2884 What you are describing is basically a categorical multiple comparisons problem.<BR /> <BR /> See<BR /> <A href="http://support.sas.com/kb/22/565.html" target="_blank">http://support.sas.com/kb/22/565.html</A><BR /> for ideas. Thu, 06 Aug 2009 12:41:12 GMT https://communities.sas.com/t5/Statistical-Procedures/post-chi-square-test/m-p/61647#M2884 Doc_Duke 2009-08-06T12:41:12Z