Statistical Methods for the Determination of Noncompliance With Occupational Health Standards
DHHS (NIOSH) Publication Number 75-159
Procedures are presented for determining if the 95% Lower Confidence Limit of the concentration of a contaminant in an occupational environmental air sample or the arithmetic average for a group of such samples exceeds an occupational health standard. The determination can be made with a known maximum probability of making an incorrect decision that a state of noncompliance exists. Recommendations are given concerning the duration of samples, the number of samples to take and the periods during the work day when the samples should be collected. When the true arithmetic average air concentration is estimated from the mean of a group of random grab samples, the log-normal distribution is assumed for the individual grab samples. When only one sample is examined or when one or several samples are taken for the entire time period for which a standard is defined the normal distribution is assumed. In the latter case, variability over time is absent but sampling and analytical errors are present.