Date: Thu, 10 Jul 2003 10:54:13 -0600
Reply-To: Susan Crawford <Susan.Crawford@CalgaryHealthRegion.ca>
Sender: "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From: Susan Crawford <Susan.Crawford@CalgaryHealthRegion.ca>
Organization: Calgary Health Region
Subject: Re: Problem selecting first cases
Content-type: text/plain; charset=us-ascii
Many thanks to Richard, Gene, and David. I really appreciate the help.
Everything works just fine now! Thanks again.
Richard Ristow wrote:
> At 09:35 AM 7/10/2003 -0600, Susan Crawford wrote:
> >We have just finished a huge study of reaction time in children, and
> >have run into trouble selecting cases for analysis.
> >The problem is that after the positive target was displayed, in some
> >trials, the child made the correct response once and in other trials,
> >the child made the response three or four times. We only want to look
> >at the first response if the child made more than one for that trial
> >when we do our analysis.
> >The data file looks like this:
> > Type of Target Time Target Came On Time Key Was Hit
> > Positive 1 45
> > Positive 1 75
> > Positive 1 95
> > Negative 100
> > Positive 300 405
> >So for the data file above, I would only want to have the first
> >record, fourth and fifth. The file for each child ends up being over
> >1000 records long, so I need to be able to use syntax to do this.
> Mantra: to get one case from several, use AGGREGATE. And, in your case,
> don't think of it as selecting the first CASE, think of it as selecting
> the smallest VALUE.
> If the three variables are named TARG_TYP, TIME_ON, and TIME_KEY, and
> the data is in the active SPSS file,
> AGGREGATE/OUTFILE=*/BREAK=TARG_TYP TIME_ON/TIME_KEY=MIN(TIME_KEY).
Susan Crawford, M.Sc.
Behavioural Research Unit
Alberta Children's Hospital
Phone: (403) 943-7364
FAX: (403) 543-9100