Welcome to my Open Notebook

This is an Open Notebook with Selected Content - Delayed. All content is licenced with CC-BY. Find out more Here.

ONS-SCD.png

handling-survey-data-with-r

R is generally very good for handling many different data types but

R has problems with survey data

This post is a stub about what packages Ive found with methods allowing to handle efficiently survey data: handle variable labels, values labels, and retrieve information about missing values

Base R:

## Not run:
require(foreign)
analyte  <- read.spss(filename, to.data.frame=T) 
varslist <- as.data.frame(attributes(analyte)$variable.labels)
# this gives a pretty useful thing to use

While I was digging around in TraMineR I found this link to Dataset, Emmanuel Rousseaux’s package for handling, documenting and describing data sets of survey data.

Code:Dataset, a package for handling-survey-data-with-r

if(!require(Dataset)) install.packages("Dataset", repos="http://R-Forge.R-project.org");
require(Dataset)
data(dds)
str(dds)
# cool
description(dds$sexe)
# excellent!

Conclusions

I’m sure there are plenty of other approaches. I’ll add them as I find them’

Posted in  Data Documentation


blog comments powered by Disqus