Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Re: RE: st: Repeated Measures ANOVA and Missing Values


From   Philip Ender <ender97@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: Re: RE: st: Repeated Measures ANOVA and Missing Values
Date   Mon, 17 Oct 2011 13:56:47 -0700

<gamirali@hotmail.com> wrote:
what is considered to be small sample size for xtmixed? from what I
understood from you xtmixed Chicago lecture with small sample size
xtmixed may cause type I error, if I am correct please advise about a
good sample size (total sample size) for xtmixed.
----------------------
Dear Amir,

The issue is not the sample size for -xtmixed- but what sample size do
you need to have confidence in the chi-square p-values.  I'm afraid
there is no simple easy answer to this question.  Most experimental
designs tend to have relatively small sample sizes and report F-ratios
and associated p-values.  If I don't have repeated measures, I use
-anova- or -regress-.  However, for repeated measures with missing
observations I use -xtmixed-.  Back when I was a S*S user I used -proc
mixed- which gives results as F-ratios with denominator degrees of
freedom.  A lot of my psychology clients use S*SS -mixed- which also
yields F-ratios and p-values.  I try to obtain the same results in
Stata, my preferred package, by estimating reasonable denominator
degrees of freedom.

-- 
Phil Ender
UCLA Statistical Consulting Group
"Money can't buy you happiness, but it can buy you a good statistical package."
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index