wassom.brome.uniformity {agridat} | R Documentation |
Uniformity trials of bromegrass
Description
Uniformity trials of bromegrass at Ames, Iowa, 1950-1951.
Usage
data("wassom.brome.uniformity")
Format
A data frame with 1296 observations on the following 3 variables.
expt
experiment
row
row
col
column
yield
forage yield, pounds
Details
Experiments were conducted at Ames, Iowa. The response variable is forage yield in pounds of green weight.
Optimum plot size was estimated to be about 3.5 x 7.5 feet.
Wassom and Kalton used two different methods to estimate optimum plot size. 1. Relative efficiency of different plot sizes. 2. Regression of the log variance of yield vs log plot size.
There are three Experiments:
Experiment E1 was broadcast seeded, harvested in 1950.
Experiment E2 was row planted, harvested in 1950.
Experiment E3 was broadcast seeded, harvested in 1951. This field contained a mixture of alfalfa and brome in about equal proportions.
Each plot was 3.5 ft x 4 ft, but the orientation of the plot is not clear.
Field width: 36 plots
Field length: 36 plots
Source
Wassom and R.R. Kalton. (1953). Estimations of Optimum Plot Size Using Data from Bromegrass Uniformity Trials. Agricultural Experiment Station, Iowa State College, Bulletin 396, page 314-319. https://lib.dr.iastate.edu/ag_researchbulletins/32/ https://babel.hathitrust.org/cgi/pt?id=uiug.30112019570701&view=1up&seq=26&skin=2021
Examples
## Not run:
library(agridat)
data(wassom.brome.uniformity)
dat <- wassom.brome.uniformity
libs(desplot)
desplot(dat, yield~col*row|expt,
flip=TRUE, aspect=1, # approximate aspect
main="wassom.brome.uniformity")
## End(Not run)