www.cmm.bristol.ac.uk/forum

I’m working with complex survey data from the Demographic and Health Surveys (DHS), where individuals are nested within clusters, which are in turn nested within countries. I’d like to estimate a multilevel model that includes country fixed effects (multi country analyses) and random intercepts at the cluster level, but I also want to account for survey weights. I have individual-level survey weights, and I understand that I need to derive approximate weights at the cluster level and then standardize them.

My main question is: how do I specify these weights in the R2MLwiN package in R? In particular, what command or syntax should I use to incorporate the weights properly

Reeth

You can find some examples of this in the following thread:

Specifying weight in R2MLwiN

This should work with the current versions of R2MLwiN and MLwiN, so you shouldn't need to install the development version mentioned there.

If you have further questions then please let us know.

Thanks for the link — really helpful. I’ll start by trying that approach with the derived cluster weights and see how R2MLwiN handles it. If I run into issues, I’ll come back with specifics.