Heteroscedastic one-way ANOVA for medians.
This function computes a one-way ANOVA for the medians. Homoscedasticity assumption not required. There shouldn't be too many ties.
med1way(formula, data, iter = 1000, ...)
formula: an object of class formula.data: an optional data frame for the input data.iter: number of iterations to determine critical value....: currently ignored.Evaluating the test statistic using the df proposed in the literature can result in the actual level being less than the nominal level, (i.e., around 0.02-0.025 when testing at the 0.05 level and the sample size is small). A better strategy is to simulate the critical value and computing the p-value accordingly, as implemented in this function.
Returns an object of class med1way containing: - test: F-value of the test statistic
crit.val: critical value
p.value: p-value
call: function call
Wilcox, R. (2012). Introduction to Robust Estimation and Hypothesis Testing (3rd ed.). Elsevier.
t1way,t1waybt, Qanova
set.seed(123) med1way(libido ~ dose, data = viagra, iter = 3000)