> With such a small data set, why not simulate some data sets with > reasonable
> effect sizes and see how an analysis performs? Krzysztof
Dear Krzysztof,
It is good idea. Would you know some R functions thatis are well suited for
this kind of simulations
_
Thanks to all for the useful suggestions -- I just wanted to report back on my
experience using the code that Dave Roberts refers to below, on his web page.
It worked beautifully.
I imagine that the utility of this approach would depend on the compositional
overlap of the new samples with t
Hi Thierry,
The multiple comparisons ran just fine but there was a ridiculous amount of
interaction combinations all of which were non-significant even though
there was a highly significant interaction term. I decided to remove test
as a variable to simplify the analysis and run separate single ex
Hi Andrew,
Please keep the mailing list in cc.
The estimates in mc are the differences of the parameter estimates (betas)
between both levels. E.g. 5.LR -1.LR = -1.168 or 5.LR = 1.LR - 1.168
summary(mc) should give you the significance of those differences. That should
work. If it doesn't, ple
With such a small data set, why not simulate some data sets with
reasonable effect sizes and see how an analysis performs? Krzysztof
On Mon, Oct 20, 2014 at 11:53 AM, V. Coudrain wrote:
> Thank you for this helpful thought. So if I get it correctly it is hopeless
> to try testing an interaction
