Simple code for a t-test with unequal variance on ranked values

wilcox_unequal_var.R

wilcox_unequal_var <- function(x, y, ...) {
  # The Wilcoxon rank sum test (Mann Whitney U test) does not perform
  # that well on data where the variances are unequal. Some
  # investigators have recommended as an alternative performing the
  # unequal variance t-test on ranked values. See
  # https://academic.oup.com/beheco/article/17/4/688/215960
  # https://psycnet.apa.org/doiLanding?doi=10.1037%2Fh0078850
  rr <- rank(c(x, y))
  nx <- length(x)
  t.test(rr[1:nx], rr[(nx+1):length(rr)], var.equal=FALSE, ...)
}