You can do a transformation on your Y, and than use the normal tools.
Using something like a median test is not advisable due to low power, and a
wilcox.test is considered a better choice (although the meaning is more
complex when observations are not symmetrically distributed).
If you wish to
Hi,
Check this link.
http://r.789695.n4.nabble.com/help-comparing-two-median-with-R-td822998.html
A.K.
- Original Message -
From: Data Analytics Corp. w...@dataanalyticscorp.com
To: r-help@r-project.org
Cc:
Sent: Thursday, July 19, 2012 4:14 PM
Subject: [R] median comparison tests
Hi,
A client has a consumption measure on each of four products. The sample
size is 75. The consumption distributions are highly skewed for each
product. He would like a pairwise comparison test of the products, much
like Tukey's HSD but using medians rather than means. Is there such a
] On Behalf Of Data Analytics Corp.
Sent: Thursday, July 19, 2012 3:15 PM
To: r-help@r-project.org
Subject: [R] median comparison tests
Hi,
A client has a consumption measure on each of four products. The
sample
size is 75. The consumption distributions are highly skewed for each
4 matches
Mail list logo
