Third, because of the natural uniform distribution of scaled rankings, data are not
skewed, kurtosis is improved, and the scaled rankings should be more robust to the
assumptions of gaussian statistical methods.
Fourth, because scaled rankings are naturally bounded, and unaffected by distance
between observations, outlier effects, if present, are markedly reduced. Similarly, to the
extent that distance data contains white noise, the sensitivity and power of statistical tests
are likely to improve.