topic Means tables from ANCOVA in SurveyReg in Statistical Procedures

Means tables from ANCOVA in SurveyReg

LindaW — Tue, 07 Feb 2012 22:47:03 GMT

Using SAS Survey:

How can I get a means table of the data resulting from an ANCOVA in proc surveyreg? I'm trying to find the variable means that result when a regression of two variables is controlled for gender (or age or another variable). I can't seem to build appropriate syntax. Any ideas?

Means tables from ANCOVA in SurveyReg

art297 — Tue, 07 Feb 2012 23:06:08 GMT

I'd think they are provided in one of the ODS tables (see: http://support.sas.com/documentation/cdl/en/odsug/61723/HTML/default/viewer.htm#a002649072.htm )

Note: Surveyreg is in the second section of the page, under the SAS/Stat procs.

Means tables from ANCOVA in SurveyReg

art297 — Tue, 07 Feb 2012 23:40:52 GMT

I probably shouldn't have posted a response as I have never used the proc. However, that said, wouldn't specifying contrasts provide what you want or, if not, proc surveymeans? Take a look at: http://support.sas.com/rnd/app/papers/survey.pdf

Means tables from ANCOVA in SurveyReg

LindaW — Wed, 08 Feb 2012 15:46:47 GMT

Thank you so much for trying to help me with this. The thing is, I need to tell SAS to provide adjusted means based on the ANCOVA and I can't see how to write that request. If I use the standard surveymeans request, the response is the unadjusted means. I can't understand how to tell SAS to provide adjusted means in SAS Survey.

Means tables from ANCOVA in SurveyReg

SteveDenham — Thu, 09 Feb 2012 12:45:34 GMT

Linda,

Have you specified an LSMEANS statement?

From what you have presented, I assume you have something like:

proc surveyreg data=have;

strata sex;

class sex;

model depvar=sex othervar sex*othervar/solution;

weight weight;

run;

Add the LSMEANS statement that looks something like the following, and it should give you marginal means, adjusted for the covariate 'othervar'. Note that these are predicted marginals for a hypothetical balanced population.

proc surveyreg data=have;

strata sex;

class sex;

model depvar=sex othervar sex*othervar/solution;

weight weight;

lsmeans sex;

run;

There are several options available to the LSMEANS statement that you might want to explore. Also, I specified a model that had different slopes for each sex in the ANCOVA--you may wish to fit a common slopes model which would not include the interaction term.

Good luck.

Steve Denham

Means tables from ANCOVA in SurveyReg

LindaW — Thu, 09 Feb 2012 13:54:25 GMT

This works perfectly! Thanks, Steve!!