search

xKDR / Survey.jl

Analysis of complex surveys
https://xkdr.github.io/Survey.jl/
GNU General Public License v3.0
50 stars 19 forks source link

apiclus2 weights #291

Closed ayushpatnaikgit closed 1 year ago

ayushpatnaikgit commented 1 year ago

" The sampling weights in apiclus1 are incorrect (the weight should be 757/15) but are as obtained from UCLA. " https://r-survey.r-forge.r-project.org/survey/html/api.html

I am uncertain if it's the case with apiclus2 as well. How can we check?

smishr commented 1 year ago

Look in R documentation of api. That is where I found that apiclus1 should be 757/15. afaik apiclus2 weights were correct? But we havent tested Survey.jl with two stage cluster samples like apiclus2 much yet, so im not sure if it is an algorithm bug or issue with dataset.

smishr commented 1 year ago

we have sorted out by removing correction for apiclus2 in runtests.jl