Conducts the modifies Levene's test for homoscedastic populations.
Usage
modlevenes.test(x, groups)
Arguments
x
Vector of residuals from a linear model.
groups
Vector of factor levels.
Value
An ANOVA table is returned with the modified Levene's test results.
}
references{
Kutner, M. H., Nachtsheim, C. J., Neter, J., and W. Li. (2005) emph{Applied linear statistical models, 5th edition}. McGraw-Hill, Boston.}
author{Ken Aho}
seealso{code{fligner.test}}
examples{
eggs<-c(11,17,16,14,15,12,10,15,19,11,23,20,18,17,27,33,22,26,28)
trt<-as.factor(c(1,1,1,1,1,2,2,2,2,2,3,3,3,3,4,4,4,4,4))
lm1<-lm(eggs~trt)
modlevenes.test(residuals(lm1),trt)
}
keyword{univar}
keyword{htest}
Details
The modified Levene's test is a test for homoscedsticity that (unlike the classic F-test) is robust to violations of normality (Conover et al. 1981). In a Modified Levene's test we calculate $d_{ij}=|e_{ij} - \tilde{e}_{i}|$ where $\tilde{e}_i$ is the ith factor level residual median. We then run an ANOVA on the $d_{ij}$'s. If the p-value is < $\alpha$, we reject the null and conclude that the population error variances are not equal.