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 at the end of May, and its replacement, statalist.org is already up and running.


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

Re: st: Using ksmirnov


From   Maarten Buis <maartenlbuis@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Using ksmirnov
Date   Wed, 5 Sep 2012 17:14:21 +0200

On Wed, Sep 5, 2012 at 4:41 PM, Morrison Hodges wrote:
> I'm trying to use ksmirnov to compare 2 distributions to determine if there are any differences in the distribution of the variables.
>
> I have a .dta file with 179 patients. Each patient is classified as being in one of 7 groups from 1 to 7, i.e., group==1, group==2, etc. I want to compare the same variable between 2 groups. All variables are already sorted, from lowest to highest.

For such a small sample size like that I would use Nick Cox's
-stripplot- (see: -ssc desc stripplot-). The power of tests like
-ksmirnov- is usually pretty bad, so in small sample sizes you are not
going to find anything that is not glaringly obvious if you just
looked at the data. On the other hand you will in all likelihood miss
lots of real an interesting patterns if you focus on such low power
tests. So, there is no added value to be got from these tests and a
real chance of a loss.

-- Maarten

---------------------------------
Maarten L. Buis
WZB
Reichpietschufer 50
10785 Berlin
Germany

http://www.maartenbuis.nl
---------------------------------
*
*   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