Learn R Programming

agridat (version 1.23)

patterson.switchback: Switchback experiment on dairy cattle, milk yield for 4 treatments

Description

Switchback experiment on dairy cattle, milk yield for 4 treatments

Usage

data("patterson.switchback")

Arguments

Format

A data frame with 36 observations on the following 4 variables.

y

response, milk FCM

trt

treatment factor, 4 levels

period

period factor, 3 levls

cow

cow factor, 12 levels

Details

There are three periods. Each cow is assigned to one treatment cycle like T1-T2-T1, where T1 is the treatment in period P1 and P3, and T2 is the treatment in period P2.

There are four treatments.

All 4*3 = 12 treatment cycles are represented.

Data were extracted from Lowry, page 70.

References

Lowry, S.R. 1989. Statistical design and analysis of dairy nutrition experiments to improve detection of milk response differences. Proceedings of the Conference on Applied Statistics in Agriculture, 1989. https://newprairiepress.org/agstatconference/1989/proceedings/7/

Examples

Run this code
if (FALSE) {

library(agridat)
data(patterson.switchback)
dat <- patterson.switchback

# Create groupings for first treatment, second treatment
datp1 <- subset(dat, period=="P1")
datp2 <- subset(dat, period=="P2")
dat$p1trt <- datp1$trt[match(dat$cow, datp1$cow)]
dat$p2trt <- datp2$trt[match(dat$cow, datp2$cow)]
                     
libs(latticeExtra)
useOuterStrips(xyplot(y ~ period|p1trt*p2trt, data=dat,
                      group=cow, type=c('l','r'),
                      auto.key=list(columns=5),
                      main="patterson.switchback",
                      xlab="First/Third period treatment", 
                      ylab="Second period treatment"))


# Create a numeric period variable
dat$per <- as.numeric(substring(dat$period,2))

# Need to use 'terms' to preserve the order of the model terms
m1 <- aov(terms(y ~ cow + per:cow + period + trt, keep.order=TRUE), data=dat)
anova(m1) # Match table 2 of Lowry
## Analysis of Variance Table
##           Df Sum Sq Mean Sq F value    Pr(>F)
## cow       11 3466.0 315.091 57.1773 2.258e-06 ***
## cow:per   12  953.5  79.455 14.4182 0.0004017 ***
## period     1   19.7  19.740  3.5821 0.0950382 .
## trt        3   58.3  19.418  3.5237 0.0685092 .
## Residuals  8   44.1   5.511

}

Run the code above in your browser using DataLab