Dear R-users,
I have 3 plant populations (fixed). Within each population there is the
same number of “families” (random) – the seed progeny of the same plant.
These families were exposed to 2 treatments (fixed) and their response was
measured (mean values for 25 seedlings per family per treatment are
presented in data table).
I would like to know if there is a significant difference in the response
of populations between the treatments (primarily the interaction term, and
the main effects as well) taking into account an important (from biological
point of view) thing that progeny of each plant (i.e. family) was exposed
to both treatments.
Taking the artificial example one could easily do it with car-package:
library(car)
dat <- data.frame(Family = 1:60, # Plant
family name
Pop = rep(c("Pop1","Pop2","Pop3"), each=20),
# Population name
Cond1 = rnorm(60, 15, 1), # obtained
values at experimental conditions 1
Cond2 = rnorm(60, 20, 1)) #
experimental conditions 2
# rearrange data
data.wide <- data.frame(Family = 1:20,
subset(dat, dat$Pop == "Pop1")[3:4],
subset(dat, dat$Pop == "Pop2")[3:4],
subset(dat, dat$Pop == "Pop3")[3:4])
names(data.wide)[2:7] <- c("Pop1.Cond1","Pop1.Cond2",
"Pop2.Cond1","Pop2.Cond2",
"Pop3.Cond1","Pop3.Cond2")
# define the structure of analysis
design <- data.frame(Pop =
rep(c("Pop1","Pop2","Pop3"), each=2),
Cond = rep(c("Cond1","Cond2")))
# define the model
mod <- lm(as.matrix(data.wide[, -1]) ~ 1)
an <- Anova(mod, idata = design, idesign = ~Pop * Cond)
summary(an)
But obviously this is not the right way to analyse this data because plant
families are nested within the populations.
So I’m struggling with how to incorporate this information into the model.
Tanks in advance for any suggestions and/or helpful links!
Vladimir.
PS. If it’ll be easier to do it with the long format of data one can run
this code:
library(reshape2)
data.long <- melt(dat, measure.vars=c("Cond1", "Cond2"),
variable.name
="Cond")
--
Vladimir Mikryukov, PhD
Institute of Plant & Animal Ecology UD RAS,
Lab. of Population and Community Ecotoxicology
[8 Marta 202, 620144, Ekaterinburg, Russia]
[[alternative HTML version deleted]]
