A.6.4.5 Fundamental error (FE)

FE is a random quantity that represents the relative difference (relative bias) between the true population parameter (e.g., mean) and the estimate of the mean obtained from sampling of bulk material. The magnitude of the FE (and its variance) is a function of CH and the mass of the collected sample (Ms).

Due to the varying composition (and distribution) of bulk material particles contained in the DU, each individual increment with the same sample support yields a different estimate of the mean. FE is addressed (i.e., reduced but not eliminated) by obtaining adequate sample mass (e.g., by collecting and combining many increments). Sampling pattern (e.g., simple random sampling) used to collect increments also plays an important role in addressing FE and reducing bias in the mean estimate; this is especially true when a DU is moderately to highly heterogeneous.