COMBINED ESTIMATES IN COLORIMETRIC DETERMINATIONS WITH HIGH VARIABILITY
In a number of colorimetric determinations used in biochemistry, the absorbances of replicates are subject to variations. The resulting variability may be time-to-time variability, measured by the intervariance, or variability of simultaneous replicates, measured by the intravariance. In such determinations two estimates of the unknown may be derived according to the standards used, one based on the absorbance of simultaneous standards, a second based on the mean absorbance of all available standards. If intravariance does not exist, the first estimate is the more precise. If intervariance does not exist, the second estimate is the more precise. If both intravariance and intervariance exist, a weighted mean of two estimates, one based on the absorbances of simultaneous standards and a second based on the mean absorbance of all the available non-simultaneous standards, is a value with an increased precision. A method for calculating this value is given.