I am trying to properly apply BRR (balanced repeated replicate) weights in the R Survey Package and I am using the National Center for Education Statistics ELS 2002 dataset. I am not sure how to figure out what to specify for the combined.weights argument, which should be true if the replicate weights already include the sample weight (which the R manual says is typically the case). How do I go about finding out if the replicate weights "include" the sample weight or not? If the combined.weights argument is true, then the scale and rscales arguments need to be specified and I am not sure how to go about that either. I have seen some examples, but they seem to differ by the dataset involved. Any advisement would be greatly appreciated.
$\begingroup$
$\endgroup$
2
-
$\begingroup$ Welcome to CV. It might help your potential respondents if you were to explain the acronyms such as "BRR." Thank you. $\endgroup$– user78229Apr 19, 2016 at 16:06
-
$\begingroup$ Balanced repeated replication $\endgroup$– Estimate the estimatorsJan 4, 2023 at 7:20
Add a comment
|