I found and loaded the "survey" package. ?rake and ?postStratify seem
promising. Are there other packages/procedures I've missed? Are there
online references that an R newbie could use to feel comfortable
applying these procedures to a survey? How about a reference discussing
the details of designing a "post survey survey" to generate additional
weights?
Thanks in advance,
Robert Farley
Metro
www.Metro.net
________________________________
From: Farley, Robert
Sent: Thursday, June 26, 2008 15:48
To: 'r-help@r-project.org'
Subject: Survey questions
First the R question. I have the results of a rather large survey
(thousands of forms, each with dozens of questions) with some existing
weights and expansion factors. I wish to add additional weighting
factors, based on new information that elements of certain variables
should appear in certain proportions. Where should I look in R to
develop what is essentially an N-dimensional balancing process. I'd
like to develop new weighting factors, without disturbing the existing
weightings/expansion.
Now the statistics question. Is this appropriate? Where can I find a
nearly entry level discussion of the statistical ramifications of this?
Is it possible (proper) to re-survey a population to obtain new weights
(for previously un-weighted responses) for an original survey?
Thanks in advance,
Robert Farley
Metro
1 Gateway Plaza
Mail Stop 99-23-7
Los Angeles, CA 90012-2952
Voice: (213)922-2532
Fax: (213)922-2868
www.Metro.net
[[alternative HTML version deleted]]