Date:         Sat, 1 Nov 2008 23:24:23 -0800
Reply-To:     David L Cassell <davidlcassell@MSN.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         David L Cassell <davidlcassell@MSN.COM>
Subject:      Re: Urgent: NOTE: 1541 observations were omitted due to
              nonpositive weights.
In-Reply-To:  <BLU132-W10D31A2FC307D5D91476E8C8230@phx.gbl>
Content-Type: text/plain; charset="Windows-1252"

Just how did you 'adjust' your weights?  Your adjusted weights could have unforseen consequences. For example, a small error might have turned some of the weights to zero by accident. Or raking could hav inadvertently turned some weights into zeroes or negative numbers (although I hope not). So it probably matters HOW you did those adjustments.

David
--
David L. Cassell
mathematical statistician
Design Pathways
3115 NW Norwood Pl.
Corvallis OR 97330

> Date: Sat, 1 Nov 2008 17:06:14 -0400
> From: sasandstats@LIVE.COM
> Subject: Urgent: NOTE: 1541 observations were omitted due to nonpositive              weights.
> To: SAS-L@LISTSERV.UGA.EDU
> 
> I have adjusted the weights for my survey data, and suddenly I am losing a lot of observations in proc surveyfreq crosstabs. None of the weights I am using are negative. I am having to adjust for cluster sampling, too.
>  
> Why does this happen? And how can it get resolved?
>  
> Here is an example of my code:
>  
> proc surveyfreq data= step1c;
> tables nocase*group*gender*bm5domain*Bcat
> /row cl chisq;
> cluster pid; 
> weight finwts; 
> run;
>  
> 
>  
> Thanks!
> D. 
> _________________________________________________________________
> When your life is on the go—take your life with you.
> http://clk.atdmt.com/MRT/go/115298558/direct/01/

_________________________________________________________________
Want to read Hotmail messages in Outlook? The Wordsmiths show you how.
http://windowslive.com/connect/post/wedowindowslive.spaces.live.com-Blog-cns!20EE04FBC541789!167.entry?ocid=TXT_TAGLM_WL_hotmail_092008