Hi Janet

Try something like this

data center;

input sex center;

datalines;

1 1

1 1

1 1

0 1

0 1

0 1

1 2

1 2

0 2

1 3

1 3

1 3

1 4

0 4

0 4

0 4

;

proc freq data=center;

table sex*center / chisq;

run;

Thank you very much for your reply. I've transposed my data to be able to run the chisq statement in that way. I imagine that I have to use the results from the  Likelihood Ratio Chi-Square since I’m interested in the ratios, correct? Also, one final question: SAS warning me that 63% of my data is missing, because of the unequal sample sizes. I wonder if this is a problem as far as the reliability of the Chi-squared results are concerned?

Janet

You can use the Chi square result. SAS warning must be not about the missing cell sizes but about  expected frequencies less than 5 in many cells. This poisons the reliability of Chi square. 

In this case you can use Fisher exact test although it is not a best-fit. You do not have many alternatives though except than Fisher exact test here.

Or you could just give the proportions and no p values.