Technical Bulletin No. 0703: A Statistical Method and Computer Program for Estimating the Mean and Variance of Multi-Level Left Censored Data Sets

Censored data sets, those containing observations that are less than the analytical limit of detection (LOD) value, are becoming more common as regulatory agencies and dischargers strive to reduce concentrations of trace contaminants in treated effluent. The presence of less than LOD or nondetect (ND) observations in a data set make estimation of the mean and standard deviation of the parent population more difficult. Prior NCASI research has examined several methods for analyzing censored data sets and of the alternatives considered, Cohen's MLE method was found to be superior in terms of bias and RMSE characteristics.