You are being redirected because this document is part of your ASTM Compass® subscription.
    This document is part of your ASTM Compass® subscription.


    The Statistics of Small Data Sets

    Published: 0

      Format Pages Price  
    PDF (224K) 14 $25   ADD TO CART
    Complete Source PDF (3.1M) 162 $55   ADD TO CART


    A primary concern in Risk-Based Corrective Action (RBCA) is the decision criteria for evaluating attainment of cleanup objectives. Carcinogenic compounds (with long modeled exposure duration) typically drive remediation at most sites. The statistic for comparison with the risk-based cleanup objective should therefore be an upper bound estimate (the “Reasonable Maximum Exposure” [RME]) for the true spatially averaged concentration. Contaminant concentrations in soil are inherently lognormally distributed. Small simulated “sample sets” selected at random from a lognormal distribution were used to test the accuracy and stability of the statistical method recommended in Superfund guidance and that recommended in RCRA SW-846 (using the log transformation). The results show that the geomean is the more accurate and stable estimator of true average exposure concentration and that the 90% Upper Confidence Limit of the geomean is a conservative statistic for comparison with risk-based standards.


    environmental data, statistical analysis, Monte Carlo, mean, variance, lognormal distribution, cleanup objective

    Author Information:

    Ball, RO
    ENVIRON International,

    Hahn, MW
    ENVIRON International,

    Committee/Subcommittee: D18.10

    DOI: 10.1520/STP13284S