Date:         Tue, 21 Nov 2006 16:04:00 -0500
Reply-To:     Statisticsdoc <statisticsdoc@cox.net>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Statisticsdoc <statisticsdoc@cox.net>
Subject:      Re: missing data
Comments: To: Kelly Sorensen <kelsoren@UGA.EDU>
In-Reply-To:  <200611211908.kALGbAjx029726@mailgw.cc.uga.edu>
Content-Type: text/plain; charset="iso-8859-1"

Stephen Brand
www.statisticsdoc.com

Kelly,

How were subjects kicked off the survey?  Were they kicked off the survey
based on specific answers to particular items?  Would it be reasonable to
select out cases with that pattern of responses to the "gateway" items?

HTH,

Stephen Brand

For personalized and professional consultation in statistics and research
design, visit
www.statisticsdoc.com

-----Original Message-----
From: SPSSX(r) Discussion [mailto:SPSSX-L@LISTSERV.UGA.EDU]On Behalf Of
Kelly Sorensen
Sent: Tuesday, November 21, 2006 2:08 PM
To: SPSSX-L@LISTSERV.UGA.EDU
Subject: missing data


I encountered a problem with a survey we administered and a number of
respondents were "kicked off" the survey before they could answer any
questions. I'd like to remove these cases from the dataset but the
instructions asked respondents to leave cases blank where they felt an
item did not apply to them. These cases I'd like to keep. There are too
many cases to reliably remove cases manually, but I'm having trouble
coming up with syntax to identify the cases where there are NO data for
the variables of interest (though we do have background data). Summing the
variables doesn't seem to work, but it SPSS doesn't add the cases with
systems missing data.

I'm working with SPSS 12.0. I've gotten some ideas from the archives, but
nothing has worked as yet.

Kelly