TRUEthen missing values are removed before computation proceeds.
sd(x, na.rm = FALSE)
varthis uses denominator $n - 1$.
The standard deviation of a zero-length vector (after removal of
na.rm = TRUE) is not defined and gives an error.
The standard deviation of a length-one vector is
varfor its square, and
mad, the most robust alternative.
sd(1:2) ^ 2