Note on the standard error of the coefficient of skewness

Based on Slade's work in 1936, it has long been considered that a skew coefficient is not reliable if it is computed from a sample of less than 140 items. This limitation is erroneous and has acted as a deterrent to the use of computed frequency distributions. A table of the standard error of the skew coefficient is presented to assess the value of a skew coefficient from any sized sample.