Date: Mon, 8 Oct 2001 09:19:49 -0400
Reply-To: John Jones <jonesj@PHARMARESEARCH.COM>
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: John Jones <jonesj@PHARMARESEARCH.COM>
Subject: Re: omitted observations in Factoranalysis?
In-Reply-To: <3BC1A191.246BC775@plan.be>
Content-Type: multipart/alternative;
Gijs,
The warning is referring to the weight statement variable wndsh7sc. 183 of
the observations are <0? The values associated with these weights cannot
be used. The "!" symbol is there because the var statement list of
variables wrap onto the next line.
John
At 02:52 PM 10/8/01 +0200, Gijs Dekkers wrote:
>--------------CFCD184267248AC7F7B68BD6
>Content-Type: text/plain; charset=us-ascii
>Content-Transfer-Encoding: 7bit
>
>Dear All,
>
>Can anyone explain this?
>
>I am running a factoranalysis on a household-dataset where the
>observations with missing variables have been left out. The relevant
>section of the log-file
>is:
>
>163 proc factor data=datin
>164 method=principal
>165 /*method=ml*/
>166 scree
>167 mineigen=0
>168 priors=asmc
>169 rotate=promax
>170 nfactors=3
>171 nocorr
>172 /*plot*/
>173 outstat=factout
>174 out=datuitF ;
>175 weight wndsh7sc ;
>176 var eqHHink_in spreken sarm cowoko garbeid gzoek is_maxong1
>176! is_wcomfort
>177 is_maxong2 is_opl1 is_opl2 is_tevr is_socint1 is_socint2
>177! is_socint3 ;
>178 run ;
>
>WARNING: 183 of 2854 observations in data set WORK.DATIN omitted due
> to nonpositive weight or truncated frequency.
>NOTE: 3 factors will be retained by the NFACTOR criterion.
>NOTE: There were 2854 observations read from the dataset WORK.DATIN.
>NOTE: The data set WORK.FACTOUT has 55 observations and 17 variables.
>NOTE: The data set WORK.DATUITF has 2854 observations and 31
> variables.
>NOTE: PROCEDURE FACTOR used:
> real time 0.22 seconds
>
>It is this warning which I do not get. Appearantly, 183 or 2854
>observations are not used, but when I look at the output-file (DATUITF,
>which has 2854 observations), there are no observations where the
>FACTOR1-FACTOR3 are missing. Does this mean that no observations were
>left out of the analysis?
>
>Moreover, does anyone know where the exclamation marks in the Log-file
>(lines 176 and177) come from? (no, they were not in the command-file).
>
>Any help would be appreciated, because warnings like this make me very
>suspicious ...
>
>Gijs
>
>--
>dr. Gijs Dekkers
>Federal Planning Bureau
>Algemene Directie
>Kunstlaan 47-49
>B 1000 Brussels, Belgium
>++32/(0)2/5077413
>fax 7373
>
>
>--------------CFCD184267248AC7F7B68BD6
>Content-Type: text/html; charset=us-ascii
>Content-Transfer-Encoding: 7bit
>
><!doctype html public "-//w3c//dtd html 4.0 transitional//en">
>Dear All,
>
>Can anyone explain this?
>
>I am running a factoranalysis on a household-dataset where the
>observations with missing variables have been left out. The relevant
>section of the log-file
>is:
>
>163 proc factor data=datin
>164 method=principal
>165 /*method=ml*/
>166 scree
>167 mineigen=0
>168 priors=asmc
>169 rotate=promax
>170 nfactors=3
>171 nocorr
>172 /*plot*/
>173 outstat=factout
>174 out=datuitF ;
>175 weight wndsh7sc ;
>176 var eqHHink_in spreken sarm cowoko garbeid gzoek is_maxong1
>176! is_wcomfort
>177 is_maxong2 is_opl1 is_opl2 is_tevr is_socint1 is_socint2
>177! is_socint3 ;
>178 run ;
>
>WARNING: 183 of 2854 observations in data set WORK.DATIN omitted due
> to nonpositive weight or truncated frequency.
>NOTE: 3 factors will be retained by the NFACTOR criterion.
>NOTE: There were 2854 observations read from the dataset WORK.DATIN.
>NOTE: The data set WORK.FACTOUT has 55 observations and 17 variables.
>NOTE: The data set WORK.DATUITF has 2854 observations and 31
> variables.
>NOTE: PROCEDURE FACTOR used:
> real time 0.22 seconds
>
>It is this warning which I do not get. Appearantly, 183 or 2854
>observations are not used, but when I look at the output-file (DATUITF,
>which has 2854 observations), there are no observations where the
>FACTOR1-FACTOR3 are missing. Does this mean that no observations were left
>out of the analysis?
>
>Moreover, does anyone know where the exclamation marks in the Log-file
>(lines 176 and177) come from? (no, they were not in the command-file).
>
>Any help would be appreciated, because warnings like this make me very
>suspicious ...
>
>Gijs
>
>--
>dr. Gijs Dekkers
>Federal Planning Bureau
>Algemene Directie
>Kunstlaan 47-49
>B 1000 Brussels, Belgium
>++32/(0)2/5077413
>fax 7373
>
>--------------CFCD184267248AC7F7B68BD6--
John T. Jones -- Phone: (910) 509-4814 -- JonesJ@PharmaResearch.com
Rise above the obstacles in life and focus on the positive...
... for the measure given is the measure you will receive....
[text/html]
| 
