Hello,
I have a big study to analyse and I am unsure of which technique to use.
I have a group of patients suffering from disease 1. This group further
divides into 4 sub groups A, B, C and D
On the other hand I have another group of patients suffering from disease 2.
This group divides into 5 sub groups E, F, G, H and I.
The aim of my analysis is to check whether there are proteins which are
significantly changed between the two different types of diseases.
For the analysis I would pool all patients suffering from disease 1 together
to obtain a single group. I would do the same for disease 2. Then I would
use either a fixed-effects or random-effects ANOVA to identify significantly
changed analytes.
When I started to read about the different types of ANOVA's I came across
the q-statistics which is used to account for the heterogeneity within a
group. So it checks whether the effect size between sub-group A, B, C and D
is approx. the same.
Does this analysis make sense or how would you analyse this kind of data? Is
there an R package which can easily deal with such situations?
Cheers,
syrvn
--
View this message in context:
http://r.789695.n4.nabble.com/fixed-effect-or-random-effect-ANOVA-model-tp4099308p4099308.html
Sent from the R help mailing list archive at Nabble.com.
