Abstract
We have implemented an efficient system of radioimmunoassay quality control using three statistical parameters: a Z-score plot for each control pool, a Z-sum plot that combines the Z-scores, and an average of normals graph. Concrete criteria were established for evaluation of assay validity. Use of these three parameters allows differentiation between random and systematic errors, detection of control sera biodegradation, and information as to sources of systematic errors. The system also allows increased reliability of results, fewer repeat assays, and rapid problem recognition and solution with a negligible increase in time spent on quality control procedures.