https://communities.sas.com/t5/Statistical-Procedures/Should-I-use-cramer-s-v-and-phi-statistics-in-SAS-to-determine/m-p/482811#M25084 <P>Both&nbsp;<SPAN>&nbsp;cramer's v and phi statistics are for contingency .</SPAN></P> <P><SPAN>If you are running a regression model, I suggest VIF or COLLIN option of proc reg.</SPAN></P> Tue, 31 Jul 2018 13:33:00 GMT Ksharp 2018-07-31T13:33:00Z https://communities.sas.com/t5/Statistical-Procedures/Should-I-use-cramer-s-v-and-phi-statistics-in-SAS-to-determine/m-p/482477#M25074 <P>Hello,</P><P>&nbsp;</P><P>I'm trying to determine collinearity between chronic disease variables. Can I use chi-square (maybe increase the significance to 0.1), and then Cramer's V or phi to determine the strength of the relationship?</P><P>Between 0-1, what is a 'safe' range where the relationship is weak and can be ignored?</P><P>Is Cramer's V interchangeable with phi?</P><P>&nbsp;</P><P>Thanks</P> Mon, 30 Jul 2018 14:22:29 GMT https://communities.sas.com/t5/Statistical-Procedures/Should-I-use-cramer-s-v-and-phi-statistics-in-SAS-to-determine/m-p/482477#M25074 Angmar 2018-07-30T14:22:29Z https://communities.sas.com/t5/Statistical-Procedures/Should-I-use-cramer-s-v-and-phi-statistics-in-SAS-to-determine/m-p/482811#M25084 <P>Both&nbsp;<SPAN>&nbsp;cramer's v and phi statistics are for contingency .</SPAN></P> <P><SPAN>If you are running a regression model, I suggest VIF or COLLIN option of proc reg.</SPAN></P> Tue, 31 Jul 2018 13:33:00 GMT https://communities.sas.com/t5/Statistical-Procedures/Should-I-use-cramer-s-v-and-phi-statistics-in-SAS-to-determine/m-p/482811#M25084 Ksharp 2018-07-31T13:33:00Z