[R-sig-ME] Comparing weighted and unweighted estimation RE: Methodological and practical issues about survey weights using lme4
pb@||ey @end|ng |rom @|r@org
Wed Jul 14 01:23:48 CEST 2021
I just wanted to confirm that WeMix is designed for survey sample weights and, consistent with good survey variance estimation practices, produces clustered standard errors. Those interested in the package can see the package's page here https://american-institutes-for-research.github.io/WeMix/index.html.
How to use the weights is an area of active research in the literature, and I'm not going to comment on that.
Paul Bailey, Ph.D.
[[alternative HTML version deleted]]
More information about the R-sig-mixed-models