9.4. One-way ANCOVA

A one-way analysis of covariance (ANCOVA) extends the one-way ANOVA to include one or more quantitative covariates. This example comes from the litter dataset in the multcomp package (see Westfall et al., 1999). Pregnant mice were divided into four treatment groups; each group received a different dose of a drug (0, 5, 50, or 500). The mean post-birth weight for each litter was the dependent variable and gestation time was included as a covariate. The analysis is given in the following listing.

Listing 9.3. One-way ANCOVA

> data(litter, package="multcomp")
> attach(litter)
> table(dose)
dose
  0   5  50 500
 20  19  18  17
> aggregate(weight, by=list(dose), FUN=mean)
  Group.1    x
1       0 32.3
2       5 29.3
3      50 29.9
4     500 29.6
> fit <- aov(weight ~ gesttime + dose)
> summary(fit)
            Df  Sum Sq Mean Sq F value   Pr(>F)
gesttime     1  134.30  134.30  8.0493 0.005971 **
dose         3  137.12   45.71  2.7394 0.049883 *
Residuals   69 1151.27   16.69
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1