topic Re: A question about the PROC FREQ in SAS Procedures

A question about the PROC FREQ

John4 — Thu, 29 Jun 2017 13:21:30 GMT

My data set look like this :
Birth Death InfantDeath Country
24.7 5.7 30.8 Albania

I am using this code for analyse my data :

PROC FREQ DATA = MYDATA;
TABLE BIRTH*COUNTRY / CHISQ;
RUN;

Cramér's V is a measure of association between two nominal variables but I dont understnd why SAS compute Cramér's V for continuous variable and categorial variable ? In the output I got 0.2236 as V cramer for these variables : Birth and Country

Re: A question about the PROC FREQ

PaigeMiller — Thu, 29 Jun 2017 13:26:12 GMT

PROC FREQ treats all variables as categorical. It doesn't know and doesn't care that you think a variable is continuous.

Re: A question about the PROC FREQ

John4 — Thu, 29 Jun 2017 13:47:43 GMT

So we have to interpret the result 0.2236 as usually ? (I mean : "weak effect", "small effect", etc...)

Or is there a method to deal with that ?

Re: A question about the PROC FREQ

PaigeMiller — Thu, 29 Jun 2017 14:11:27 GMT

@John4 wrote:

So we have to interpret the result 0.2236 as usually ? (I mean : "weak effect", "small effect", etc...)

Or is there a method to deal with that ?

If I am understanding the question properly (and I may not be), you would not try to interpet a calculations of Cramer's V in this situation as it is meaningless.

Re: A question about the PROC FREQ

John4 — Thu, 29 Jun 2017 14:16:05 GMT

Yes I think It does not make sense