Skip to content

Problems with weight

6 messages · Anthony Damico, Pablo Menese, Milan Bouchet-Valat

3 days later
#
Le mardi 27 novembre 2012 ? 18:33 -0300, Pablo Menese a ?crit :
You were advised to use svyglm(), not glm(). It's usually considered
polite to read carefully the anwsers you get to your questions...


Regards
#
Le mercredi 28 novembre 2012 ? 14:20 -0300, Pablo Menese a ?crit :
No, I had not read this message when I wrote the mail because you sent
two completely different messages in two different threads at about the
same time. As you can see, I was replying to the other message, which
did only mention glm().
Have you read ?svydesign? It has a "variables" argument that you can use
to specify the variables you need to include in the design object. The
documentation says:
variables: Formula or data frame specifying the variables measured in
          the survey. If ?NULL?, the ?data? argument is used.

So if you want to include all variables from your original data set,
pass it as the "data" argument, and that's all.


But first, stop using attach(), it creates confusion and is probably the
reason why you did not think of passing your data.frame object to
svydesign().


Regards