survey data analysis

[willdebras]

Maybe a niche interest, but @tomasokal and I have long pondered creating better online resources, e.g. a bookdown, for survey analysis. Lumley's site has a lot, but it is pretty old and isn't the most navigable. If anybody is interested in contributing a couple pages of a bookdown, brainstorming R packages we could create or improve upon, or developing shiny apps related to survey analysis, you would be in good company.

[tomasokal]

Here is the website on Survey Analysis by Lumley:

• http://r-survey.r-forge.r-project.org/survey/

Here is a vignette on srvyr (tidyverse port of {survey}):

• https://cran.r-project.org/web/packages/srvyr/vignettes/srvyr-vs-survey.html

[tripleplusone]

I'm also interested in this! An idea I had is something that takes the output of a weighting procedure (like raking) and allows you to grab the results via the broom package. Haven't really thought out all the details, but I know weighting is something common that can be done a few different ways (here's a pretty decent example).

[willdebras]

This is a really great idea @tripleplusone. Generally, outfitting broom with more survey methods in general is an awesome idea. Alex Hayes and co. would certainly welcome PRs if documented.

[emilyriederer]

Potentially of interest is new pewmethods package that Pew just released with functions they commonly use for survey analysis. Article:

Weighting Survey Data with pewmethods