how to get adjusted means in Stata complex survey (e.g. svy commands)

[Richard Larouche]

Hi,

I am running generalized linear models using the stata (version 13) complex survey commands and my analyses use survey weights and bootstrap weights. I would like to know how to get adjusted means values for the 3 levels of my independent variable (which is a categorical variable) adjusted for the confounders (age, gender and parental education - all entered as categorical variables).

Thanks in advance!

Richard

 

[bukharin]

Use -margins-, the same as you would after running a standard model. See -help svy_postestimation-

 

[Richard Larouche]

Thanks for your response! I have tried this, but Stata gave me an error message (option invalid) with the bootstrap weights.

One solution that a statistician suggested me would be to report only the coefficients from the GLM with their 95% confidence intervals and p values.