Content area
Full Text
Abstract
Peirce's criterion is a rigorous method based on probability theory that can be used to rationally eliminate outlying or spurious data from a set of experimental measurements. Currently, a method known as Chauvenet's criterion is used in many educational institutions and laboratories to perform this function. Although Chauvenet's criterion is well established, it makes an arbitrary assumption concerning the rejection of the data, but Peirce's criterion does not. In addition, Chauvenet's criterion makes no distinction between one or several suspicious data values, but Peirce's criterion is a rigorous theory that can be easily applied to several suspicious data values. This paper describes the application of both Peirce's and Chauvenet's methods to a set of data measurements and shows the different results returned by each method.
Introduction
Peirce's criterion has been buried in the scientific literature for approximately 150 years, and is virtually unknown today in the scientific community. In its place, Chauvenet's criterion is commonly used for rational elimination of "outlier" data by government laboratories such as the Environmental Protection Agency, the U.S. Army Corps of Engineers, the Agency for Toxic Substances and Disease Registry, and the Institute for Telecommunication Sciences; by industries such as Boeing and Sikorsky; by foreign laboratories such as the Laboratoire National Henri Becquerel and the Joint Astronomy Centre; and by universities such as the University of Michigan, Texas A&M, the University of California, Vanderbilt, the University of Alberta, and Ohio State.1
Elimination of data "outliers" is useful for anyone working in industry or in an educational institution where statistical information concerning product runs or experimental data is of interest. In an engineering, technology, or science program, participants in chemistry, physics, and engineering laboratory courses often need to rationally eliminate spurious values from sets of collected data.
In the BSME program at the University of New Haven, Chauvenet's criterion has been used for many years in the instrumentation, fluid, and thermal laboratory courses. Typically, students take several measurements of a quantity such as pressure at one setting, meaning the experimental conditions are maintained at the same level. Assuming the systematic errors are negligible, each measurement will vary slightly due to random errors such as varying flow rate or misread instrument values, Often, however, one or two datum points...