LISTSERV at the University of Georgia
Menubar Imagemap
Home Browse Manage Request Manuals Register
Previous messageNext messagePrevious in topicNext in topicPrevious by same authorNext by same authorPrevious page (August 2001)Back to main SPSSX-L pageJoin or leave SPSSX-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:   Mon, 27 Aug 2001 14:18:59 -0400
Reply-To:   "Cohen,Diane" <dc42@EXCHANGE1.DREXEL.EDU>
Sender:   "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:   "Cohen,Diane" <dc42@EXCHANGE1.DREXEL.EDU>
Subject:   MISSING VALUES
Comments:   To: Mark Casazza <casazzam@adelphi.edu>
Content-Type:   text/plain; charset="windows-1252"

Dear folks (especially Mark!). I thought when one declared a missing value, this value would be excluded from frequencies and statistical tests, as Mark suggests. But, where I should have had maximum total scores of 80, I had many over 80, up to (horror of horrors) 726, which led me to believe that 99's were being added in. Bottom line, I made a mistake, which found and corrected. Now my results appear content and settled. But for my peace of mind, is it truly the case that if one declares a missing value, it is missing forevermore from any statistical tests? dgc

-----Original Message----- From: Mark Casazza [mailto:casazzam@ADELPHI.EDU] Sent: Monday, August 27, 2001 1:14 PM To: SPSSX-L@LISTSERV.UGA.EDU Subject:

Hi Diane,

I'm not sure I understand what you are asking. With the syntax you provided, a missing value (user missing as well as system missing) in any of your source variables will cause your computed variable to be system missing. Thus, if you have 9 and 99 in the original variables defined as user missing the computed variable will be system missing. If what you're asking is how to make 9 and 99 missing in the source variables, that would be:

missing values rand86r to rand95r (9,99).

as long as those variables are contiguous. If they aren't contiguous you would need to specify all of them. If what you are asking is to make the computed variable missing if the sum of the 10 source variables is either 9 or 99 then the syntax would be:

missing values randsum (9,99).

FYI, if the variables are contiguous shorter syntax would be:

compute randsum = sum.10(rand86r to rand95r).

where the ".10" specifies that 10 valid values are required--which makes it equivalent to the v1 + v2 +... version.

HTH,

Mark V. Casazza Office of Research, Assessment & Planning Adelphi University Levermore Hall, #307 Garden City, NY 11530

phone: (516) 877-3231 fax: (516) 877-3237 email: casazzam@adelphi.edu

"Those who cast the votes decide nothing. Those who count the votes decide everything." -Joseph Stalin ----- Original Message ----- From: "Cohen,Diane" <dc42@EXCHANGE1.DREXEL.EDU> Newsgroups: bit.listserv.spssx-l To: <SPSSX-L@LISTSERV.UGA.EDU> Sent: Monday, August 27, 2001 12:49 PM

> How do I add missing value = 9, 99 to the following syntax? > > COMPUTE randsum = rand86r + rand87r + rand88r + rand89r + rand90r + rand91r > + > rand92r + rand93r + rand94r + rand95r . > EXECUTE . > > Thanks > Diane D'Newbie > Philadelphia, PA > One click helps eliminate landmines > http://www.clearlandmines.com/ >


Back to: Top of message | Previous page | Main SPSSX-L page