Date:         Thu, 14 May 1998 09:48:15 -0500
Reply-To:     "Geddes, David" <geddesd@FLEISHMAN.COM>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@UGA.CC.UGA.EDU>
From:         "Geddes, David" <geddesd@FLEISHMAN.COM>
Subject:      Re: recoding variables
Comments: To: "Ann Jacobson, PhD, RN" <ajacobso@KENT.EDU>
Content-Type: text/plain

Ann,
 
from your description, I see two options:
 
(1) It sounds as if you are recoding into the same variable.  You want
to use the option of recoding into different variables.  That way, you
preserve your initial data.
 
(2) A second option would be to recode your initial variables into the
same variables so that response 1, for example, is always the "correct"
response.  However, you would also have to change your value labels to
match.
 
> ----------
> From:         Ann Jacobson, PhD, RN[SMTP:ajacobso@KENT.EDU]
> Sent:         Thursday, May 14, 1998 12:32 PM
> To:   SPSSX-L@UGA.CC.UGA.EDU
> Subject:      recoding variables
>
> A colleague asked me the following question. I post it to the list
> with
> her permission. Will forward any feedback to her.
>
> Thanks, Ann
>
> pjd@nowonline.net wrote:
> >
> > Ann. The SPSS problem is related to recoding variables into a new
> > variable.
> > When I take an existing variable with 4 possible responses eg
> 1,2,3,4
> > and I want to recode so only one of the responses is identified as
> > "correct" eg if item 2 from one variable is correct the recode would
> > look like 1=0, 2=1, 3=0, 4=0. That works fine. When I go to the next
> > variable and the "correct response" is eg 3 then I recode 1=0, 2=0,
> 3=1,
> > 4=0.  This recode works fine but then once it is executed the
> previous
> > variable that was recoded is now recoded with the second variable
> > recode. Any ideas?
> >
> > Peggy
>