Simple procedures for calculating confidence intervals around the sample mean and exceedance fraction derived from lognormally distributed data.
Appl Occup Environ Hyg 1997 Feb; 12(2):132-142
Simplified methods for the determination of accurate estimates of the 95% upper and lower confidence limits for the simple arithmetic mean and for the sample exceedance fraction for lognormally distributed exposure data were described. Calculation of the arithmetic mean was based on an adaptation of Land's procedure for the determination of exact confidence intervals around the mean of lognormally distributed data. Odeh and Owen confidence limits for a proportion in one tail of the normal distribution were used to determine the exceedance fraction. Ninety percent confidence intervals determined from the two 95% confidence limits were indicative of the degree of uncertainty in the estimates of the mean and exceedance fractions which was a function of both the inherent environmental variability in exposures and in sample size. These calculations were illustrated using real data of exposure of a worker to welding fumes on 5 consecutive days. A system of approximation functions for the automatic calculation of Land's confidence limits for the mean of lognormal data and for Odeh and Owen confidence limits for the exceedance fraction was presented in two appendices.
NIOSH-Author; NIOSH-Publication; Mathematical-models; Statistical-analysis; Exposure-limits; Threshold-limit-values; Welders
Paul Hewett, National Institute for Occupational Safety and Health, 1095 Willowdale Road, Morgantown, West Virginia 26505-2888
Applied Occupational and Environmental Hygiene