|Date: ||Wed, 10 Jan 2007 23:18:40 -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: Missing Values and Outliers|
|Content-Type: ||text/plain; format=flowed|
>Missing Values issue:
>So far, there are no guidelines for how much missing data can be
>tolerated for a
>sample of any given size (Tabachnick & Fidell, 1996).
>Is it correct to analyse the potentially meaningful influences of
>only when the amount of missings exceeded 7%???
>I have seen the scales (item by item) in my study and the % of missings
>not exceeds 1,3 %.
>I have notice several techniques to scan for outliers (steam-and-leaf;
>which strategy shoulf i use? Step-by-step.
That 7% is one of the made-up numbers that you see for rules of
thumb. I would want to check the effect of missing values. Period.
If you only have 1.3% missing values across your entire data set, then
the effects may be negligible. But trying your analyses with PROC
MI to get multiple imputation for your data can't hurt. And you
can beat someone over the head with it at some point. :-)
As for outliers, it depends on what your data are like, and what
you are trying to do with them. Univariate techniques may be
useful, and they may be a big waste of time (at best.) So, what
are you trying to do with your data that requires this work?
David L. Cassell
3115 NW Norwood Pl.
Corvallis OR 97330
Type your favorite song. Get a customized station. Try MSN Radio powered
by Pandora. http://radio.msn.com/?icid=T002MSN03A07001