[R] SPSS long variable names

[Orvalho Augusto]

Impressive!

PSPP produces the sav fine. The sav file produced by the PSPP is seen
by SPSS (16) with the long names.

I attach my sav produced by the PSPP. And I start to suspect my
foreign package version:
I am using 0.8.26-1 from the r-cran-foreign a Ubuntu 8.10 package. Can
you tell me your version? I am updating mine.

Thanks guys.

Caveman


On Thu, Oct 15, 2009 at 9:02 PM, Robert  Baer <rbaer at atsu.edu> wrote:
>>>>>> The problem is the limit of 8 characters long on variable
>>>>>> names.[in read.spss??????]
>
> I just created a file in SPSS 17 and saved it in standard format (.sav) The
> file had four unique variable NAMES which were much longer than 8 characters
> (although not 64).  I entered two rows of data and saved the file again. The
> "variable view" in SPSS looked like the following for the 4 variables I
> created.  (unique in the first postition after the first 9)
>
> a12345678901234567890987654321098765432011111111111111 Numeric 8 2
> a1234567890 None None 14 Right Scale
> a1234567899123456789098765432109876543201 Date 9 0 b1234567890 None None 14
> Right Scale
> a1234567898123456789098765432109876543201 String 8 0 c1234567890 None None
> 14 Left Nominal
> a1234567897123456789098765432109876543201 Numeric 8 2 d1234567890 None None
> 14 Right Scale
>
> I now started R and used read.spss() from the foreign package using the
> following code"  The data frame I read in seemed to reflect the original
> file with respect to column names:
>
>> library(foreign)
>> c=read.spss(file.choose(),to.data.frame=TRUE)
>> c
>
>  a12345678901234567890987654321098765432011111111111111
> a1234567899123456789098765432109876543201
> a1234567898123456789098765432109876543201
> 1                                                   33.3 13452825600
>                          a cat
> 2                                                   22.2 13463280000
>                          dog
>  a1234567897123456789098765432109876543201
> 1                                     32165
> 2                                 321654987
>
> It appears these long variable names read in just fine.  Thus, I would ask
> caveman whether the problem is really an R problem at all, or is it a
> problem with PSPP writing (in which case this would be the wrong list to be
> discussing it on)?
>
> R
>