How to statistically comapre difference between two groups with the difference between another two groups

[ZC]

Hi

I am running the following PROC MIXED precedure to examine difference between 2 groups with the difference between another 2 groups.

PROC MIXED;

     CLASS group;

     MODEL Y = group + x1 + x2;

RUN;

Categorical variable "group" has four levels: 1, 2, 3, 4.

I would like to compare the difference in Y between group 1  and 2 to the difference in Y between group 3 and 4. How can I do it?

Thanks,

[PGStats · Accepted solution]

Do not use plus signs in the MODEL statement. For the comparison, use the CONTRAST statement:

CONTRAST 'Difference of differences 1 - 2 - (3 - 4)' group 1 -1 -1 1;

PG

[ZC]

Thank for answering my question. It is very helpful.

I meet another challenge:

Another confounder (e.g., x3) in group 1 and group 3 is physiological higher than in group 2 and group 4. So I cannot include x3 as a independent variable in the model since putting x3 in the model will allow x3 equal in the four groups. To make the comparison of Y between (1-2) - (3-4) reasonable, I have to allow x3 comparable between group 1 and group3 and between group 2 and group 4.

How can I achieve it?

Thanks,