[R] Using SPSS Labels

Prof Brian Ripley ripley at stats.ox.ac.uk
Wed Dec 24 08:21:27 CET 2008


1) We didn't get the attachment.  The allowed types are given in the 
posting guide and are pretty restrictive: the suggestion is to put binary 
files on the web for download.

2) We don't have the version information asked for in the posting guide. I 
would only expect this to work in foreign 0.8-29 or 0.8-30.  If you are 
using one of those, please send me the file directly and I will take a 
look at what is happening.

3) Since you have the value labels attribute you can use it to change the 
second column into a factor.  Again, without more details I cannot tell 
you exactly what is needed.


On Tue, 23 Dec 2008, Andrew Choens wrote:

> I am trying to import a SPSS.sav file into R. The attached file is not
> technically the file I am trying to import, but does replicate my
> problem. The actual file is much too large to attach. No matter what I
> do, I can not get R (base or Hmisc) to apply the value labels in
> the .sav file to the dataframe created in R. Here's the code that I am
> using.
>
> maine <- spss.get("test.sav")
> # or
> maine2 <- read.spss("test.sav", read.value.labels=TRUE)
>
> When I try to import the file, the value labels are not assigned to the
> rows. This is what I get.
>
>   ID GENDER
> 1   1      1
> 2   2      2
> 3   3      1
> 4   4      2
> 5   5      1
> 6   6      1
> 7   7      1
> 8   8      2
> 9   9      2
> 10 10      1
> 11 11      3
>
> In the .sav file, 1 = Men 2 = Women 3 = user assigned missing.
>
> The variable values are attached as a value.labels attribute. If I
> remove row # 11 (gender = 3), I can import the file as I expect.
>
>   ID GENDER
> 1   1      Men
> 2   2      Women
> 3   3      Men
> 4   4      Women
> 5   5      Men
> 6   6      Men
> 7   7      Men
> 8   8      Women
> 9   9      Women
> 10 10      Men
>
> Given all of this: How can I import a .sav file with user assigned
> missing values correctly.
>
> If this is not possible, what is the best way to use the value.labels
> attribute when I make a table with table(Gender).
>
> Thanks.
>
> -- 
> Insert something humorous here.  :-)
>

-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595



More information about the R-help mailing list