agridat (version 1.12)

steel.soybean: Soybean yields in North Carolina

Description

Soybean yields in a multi-environment trial in North Carolina.

Format

A data frame with 108 observations on the following 4 variables.

gen

Genotype factor, 12 levels

loc

Location factor, 3 levels

block

Block factor, 3 levels

yield

Yield, grams per plot

Details

The experiment was planted as an RCB with 3 blocks at each location.

References

Mick O'Neill, 2010. Anova and REML: A Guide to Linear Mixed Models In An Experimental Design Context.

Examples

# NOT RUN {
data(steel.soybean)
dat <- steel.soybean
require(lattice)
dotplot(gen~yield|loc, dat, main="steele.soybean - yield by location")

if(require(lme4)){
# REML analysis as in O'Neill, p. 130.
m1 <- lmer(yield ~ loc + (1|gen) + (1|gen:loc) + (1|loc:block), data=dat)
# Are gen:loc effects significant?  No.
m2 <- lmer(yield ~ loc + (1|gen) + (1|loc:block), data=dat)
anova(m1,m2)
# h^2 deviations
ranef(m2)$gen
}

# }