When performing flow-frequency analysis, the mean, standard deviation, and skew coefficients are estimated from a sample of annual maximum series. The skew coefficient estimated from the sample, referred to as the station skew, is difficult to estimate accurately as it is sensitive to outliers. The estimated skew coefficient for short records may not provide an accurate representation of the true population skew coefficient.
Regional skew, also referred to as generalized skew, is an estimate of skew based on multiple samples from a homogeneous population. Regional skew can improve the probability estimates in a flow-frequency analysis by combining data from multiple gages to increase the effective sample size.