On Sat, 14 Aug 2010, ted.hard...@manchester.ac.uk wrote:
Hi Thomas,
I'm not too sure about your interpretation. Consider:
It seems hard to interpret The formula interface is only applicable for the
2-sample tests. any other way
Johannes' original query was about differences when there
are
On 14-Aug-10 16:07:14, Thomas Lumley wrote:
On Sat, 14 Aug 2010, ted.hard...@manchester.ac.uk wrote:
Hi Thomas,
I'm not too sure about your interpretation. Consider:
It seems hard to interpret The formula interface is only applicable
for the 2-sample tests. any other way
Johannes'
Hello all,
due to unexplained differences between statistical results from
collaborators and our lab that arose in the same large proteomics
dataset we reevaluated the t.test() function. Here, we found a weird
behaviour that is also reproducible in the following small test
dataset:
Thanks for the clear example. However, if there is a bug it is only that
t.test.formula doesn't throw an error when given the paired=TRUE option.
The documentation says The formula interface is only applicable for the 2-sample
tests., but there probably should be an explicit check -- I
Thank you for the fast reply! Although I have read the help page for
t.test over and over again I have obviously overlooked the relevant
sentence. The workaround that I have planned seems to be the
correct use.
Thanks again,
J. W. D.
At 15:31 Uhr -0700 13.08.2010, Thomas Lumley wrote:
Hi Thomas,
I'm not too sure about your interpretation. Consider:
set.seed(54321)
X - rnorm(10) ; Y - rnorm(10)
XY - c(X,Y); group-c(rep(0,10),rep(1,10))
t.test(X,Y,paired=TRUE)
# Paired t-test
# data: X and Y
# t = -1.5265, df = 9, p-value = 0.1612
# 95 percent
(Ted Harding) wrote:
Johannes' original query was about differences when there
are NAs, corresponding to different settings of na.action.
It is perhaps possible that 'na.action=na.pass' and
'na.action=na.exclude' result in different pairings in the
case paired=TRUE. However, it seems to me
7 matches
Mail list logo