How to analyze survey data with three levels of data in SAS?

[Anna Johanson]

We analyze data from a European survey (34 countries ) for quality of life , the sampling procedure is described as a " multi-stage stratified and clustured sampling design " . Our outcome variable is binary.
There are three levels of data to be taken into account (individuals, regions, countries). I have read a few things about these procedures, some suggesting that proc glimmix should be used ( since one can take into account the three levels of data ), but also others stating that one should apply proc survey logistic, since the parameters Estimates might otherwise be biased . However in this procedure, one cannot consider the different levels of data. I am not an expert in SAS, so I am unsure which proc would give me unbiased estimates.
Can you make a recommendation, which procedure should be used? Or another procedure?

Thank you for your help in advance!

 

[noetsi]

It really depends on which statistical method you want to use. For instance you might run multi-level models, various types of regression etc and all of these will be in different PROCS. I have run stratified samples in PROC LOGISTICS and assume it would haver worked just as well in PROC REG had I had an interval DV. Cluster sampling is something I have not worked with in SAS.

I think you need to start with the type of analysis you want to do, and then work from there.

 

[Anna Johanson]

We want to run logistic regressions, which is possible both in proc surceylogistics and in proc glimmix...
I am just wondering what is more important to take into accoutn the three loevels of data, altough I am juzst looking at one, but the analyis will take into account that the data is organized in three levels (which would be possible in proc glimmix) or to take into account that we are analyzing survey data, that has special properties, where proc surveylogistic is made for...