ABSTRACT
Error is a cause of variability in measurements. This Gaussian principle is so fundamental to psychological research that even a tremor of disbelief threatens to topple the vast edifice of statistical methods balanced so precariously upon it. In spite of such an unmistakable danger, perhaps it is time to question the traditional account of variability in psychological measurements. Among other things, this challenge to orthodox theory will provide a different view of the source of variability and, as an example of its utility, produce a new foundation for, and interpretation of, Weber's Law and Weber's Constant.
