Dear R-enthusiasts,
I am trying to do a Cochran-Armitage test for trend in R. After consulting
google I found Torsten Hothorn's remark that the 'coin' library
could be
used.
lungtumor <- data.frame(dose = rep(c(0, 1, 2), c(40, 50, 48)),
tumor = c(rep(c(0, 1), c(38, 2)),
rep(c(0, 1), c(43, 7)),
rep(c(0, 1), c(33, 15))))
table(lungtumor$dose, lungtumor$tumor)
### Cochran-Armitage test (permutation equivalent to correlation
### between dose and tumor), cf. Table 2 for results
independence_test(tumor ~ dose, data = lungtumor, teststat = "quad")
(http://tolstoy.newcastle.edu.au/R/help/05/11/16601.html).
In Peter Dalgaards Introductory statistics with R, a similar test for trends
in proportions (prop.trend.test) is described (page 134).
I wonder whether prop.trend.test and indepence_test are actually
similar.>From http://www-stat.stanford.edu/~rag/stat141/exs/nov17 and the example
above one would think so.
Since my dataset does not contain only integers (see below) I cannot get
Torsten's example to work with my data. I would appreciate some help, and
would like to excuse in advance if the answer is trivial. I am a noob, I
know.
for example:
observed 5.5 5.0 5.5 3.5 11.0 9.5 16.0 21.5 15.0 24.5
expected 11.7 11.7 11.7 11.7 11.7 11.7 11.7 11.7 11.7 11.7
Thanks in advance,
MHB
[[alternative HTML version deleted]]
