| Date: | Mon, 10 May 2004 14:25:14 -0400 |
|---|
| Reply-To: | "Jed J. Teres" <jjt2002@med.cornell.edu> |
|---|
| Sender: | "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU> |
|---|
| From: | "Jed J. Teres" <jjt2002@med.cornell.edu> |
|---|
| Subject: | Re: selecting cases |
|---|
| In-Reply-To: | <003401c436b8$77bbdf60$6501a8c0@700XL> |
|---|
| Content-Type: | text/plain; charset="us-ascii"; format=flowed |
|---|
Listers-
Thanks for your help. Your comments pointed me in the right direction. In
the end, I used the LEAD command, which is like LAG, but sets variables
back instead of forward.
Jed
At 01:58 PM 5/10/2004, Bob Walker wrote:
>SELECT IF LAG(EVENT)=1
>
>or similar if your variable is different.
>
>Bob Walker
>Surveys & Forecasts, LLC
>www.surveys-forecasts.com
>
>
>-----Original Message-----
>From: SPSSX(r) Discussion [mailto:SPSSX-L@LISTSERV.UGA.EDU] On Behalf Of Jed
>J. Teres
>Sent: Monday, May 10, 2004 1:42 PM
>To: SPSSX-L@LISTSERV.UGA.EDU
>Subject: selecting cases
>
>Hello all.
>
>I have an observation-level data set and I need to select observations in
>which a certain event occurs as well as the preceding interval.
>
>Selecting the event intervals is easy- I can just say, "select if (event eq
>1)." How can I get the case immediately preceding this one?
>
>Thanks,
>
>Jed Teres
Jedediah Teres
Data Analyst
Weill Medical College of Cornell University
Department of Psychiatry
525 East 68th Street
New York, NY 10021
(212) 746-3874
|
