From 6415437f3b69783aa2300c85dd04a28461457fc2 Mon Sep 17 00:00:00 2001 From: Daniel Date: Wed, 18 Dec 2024 11:46:40 +0100 Subject: [PATCH] docs --- R/rescale_weights.R | 6 ++++++ man/rescale_weights.Rd | 6 ++++++ 2 files changed, 12 insertions(+) diff --git a/R/rescale_weights.R b/R/rescale_weights.R index de5d1874e..61d2f2738 100644 --- a/R/rescale_weights.R +++ b/R/rescale_weights.R @@ -68,6 +68,12 @@ #' divided by the squared mean of the weights. The scales sample weights are #' then divided by the design effect. #' +#' Some tests on real-world survey-data suggest that, in comparison to the +#' Carle-method, the Kish-method comes closer to estimates from a regular +#' survey-design using the **survey** package. Note that these tests are not +#' representative and it is recommended to check your results against a +#' standard survey-design. +#' #' @references #' - Asparouhov T. (2006). General Multi-Level Modeling with Sampling #' Weights. Communications in Statistics - Theory and Methods 35: 439-460 diff --git a/man/rescale_weights.Rd b/man/rescale_weights.Rd index 65b4ec6cd..907f7314d 100644 --- a/man/rescale_weights.Rd +++ b/man/rescale_weights.Rd @@ -85,6 +85,12 @@ which means the sum of all weights equals the sample size. Next, the design effect (\emph{Kish 1965}) is calculated, which is the mean of the squared weights divided by the squared mean of the weights. The scales sample weights are then divided by the design effect. + +Some tests on real-world survey-data suggest that, in comparison to the +Carle-method, the Kish-method comes closer to estimates from a regular +survey-design using the \strong{survey} package. Note that these tests are not +representative and it is recommended to check your results against a +standard survey-design. } } \examples{