Re: removing ?formatted missing values #7

First off - let me say I joined this list about 2 weeks ago- and am enjoying
the stimulating discussions.  Its so often that we not able to keep up with
SAS- and the dynamic, spirit and advice that is being provided by the guru's
on this list is awesome.  Great work!!!

Thank you David for your sound advice.  I have some familiarity with survey
data (only been at it for
about 10 years now ;))- I'm going with IVEWARE- because it is a really good
imputation program- and also the National Center for Health Statistics used
it for their income imputations for the National Health Interview Survey-
and the survey I'm working is the child component of that- so there is some

I'm only a recent convert to imputation.

Thanks much,

-----Original Message-----
From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU]On Behalf Of
David L Cassell
Sent: Tuesday, November 29, 2005 3:39 PM
Subject: Re: removing ?formatted missing values

mah-j@statworks.com wrote:
>I'm working with a national survey that has the missing values set as .L
>.M- (I'm not sure whether this is some kind of a format as these show up as
>. missing in frequencies etc.)- is there a general statement that I can use
>to remove these from the 100+ variables in the dataset- such as format
>for real formats?
>The reason I need to do this is that I'm running an imputation using
>(SAS callable procedure)- and for some reason- this program doesn't like
>.L and .M's
>ps: Any users of IVEWARE on this list- willing to humor a few questions?

[1]  If you have a "national survey", I hope you're going to use survey
methods for analysis.  Since you're working with IVEware, I assume you have.

[2]  You can use PROC MI in SAS 9 for multiple imputation as well, and then
you can get more help from this list.  PROC MI *does* know how to handle
what we call "special missing values".

[3]  The purpose of special missing values is to provide additional
meta-data to
the user.  It's not only a missing value, it's also a floorwax!  Sorry, I
couldn't resist
the old SNL joke.  It is not only a missing value, it is also a descriptor
for the cause
of the missing-ness, or the consequences of the missing-ness.  So look up
meaning of these special missing values, because they may need to be treated
differently than other missings.  That may impact the imputation, as well as
later analyses.

[4]  You can always ask IVEware quesitons here.  You may not find anyone
with a
good answer, but you can ask...

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

Don�t just search. Find. Check out the new MSN Search!
msoobader (65)
11/30/2005 2:51:06 PM
comp.soft-sys.sas 142828 articles. 3 followers. Post Follow

0 Replies

Similar Articles

[PageSpeed] 8