Hello there,
I am trying to run an ANOVA model using a non-Standard F ratio. Imagine that
the treatments (treatments 1 & 2) are applied to the row not to individual
samples. Thus the row is the experimental unit. Therefore my error term in
my ANOVA table should be the error associated with with row.
The question is how do I check the assumptions of an ANOVA model when I have
a non-standard F ratio?
For this type of model I would normally use plot(model) to examine the
residuals. However this doesn't seem to work and I expect that R is looking
for residuals that don't exist. Is there some option I can change on the
plot command?
Sorry if this is simple but searching for this answer was a little difficult
as plot() has many uses. Below is an example. I am using R 2.10.1 and Ubuntu
9.04.
Thanks in advance!
Sam
x <- runif(48, 2, 70)
data <- data.frame(x)
data$treat1 <- factor(c("ONE", "TWO", "THREE"))
data$treat2 <- factor(c("PRUNED", "UNPRUNED"))
data$row <- factor(1:12)
model <- with(data, aov(x ~ treat1 + treat2 + treat1*treat2 + Error(row)))
plot(model)
Error in plot.window(...) : need finite 'xlim' values
In addition: Warning messages:
1: In min(x) : no non-missing arguments to min; returning Inf
2: In max(x) : no non-missing arguments to max; returning -Inf
3: In min(x) : no non-missing arguments to min; returning Inf
4: In max(x) : no non-missing arguments to max; returning -Inf
--
*****************************************************
Sam Albers
Geography Program
University of Northern British Columbia
3333 University Way
Prince George, British Columbia
Canada, V2N 4Z9
phone: 250 960-6777
*****************************************************
[[alternative HTML version deleted]]
