Rejection of Data©
Kevin Lehmann
Department of Chemistry
Princeton University
Princeton, NJ 08544
United States
mail to: Lehmann@princeton.edu



Abstract
In this document the author discusses the topic of how to 'throw out' data in an objective way. In a systematic development of the topic the author generates a set of data using a Gaussian distribution and then analyzes the distribution using standard statistical methods to demonstrate how these methods produce the properties of the parent distribution from which the data originated. Chauvenet's criteria is used to decide which data points are to be rejected and the consequences of the rejection of some data on the properties of the sample distribution. The document goes on to discuss the conditions under which the Q-test is used. The exercises in the document give students opportunities to practice the concepts. The document provides a numerical example of how statistical methods can reduce the errors in information extracted from measurements with real, as opposed to Gaussian, noise characteristics. This document would be valuable for anyone wishing to learn more about error rejection methods. Other Mathcad documents by this author, soon to appear in this collection, will address other aspects of numerical methods in chemistry.
Keywords
Audiences: Upper-Division Undergraduate, Graduate Education / Research
Pedagogies: Computer-Based Learning
Domains: Analytical Chemistry, Laboratory Instruction, Physical Chemistry
Topics: Chemometrics, Mathematics / Symbolic Mathematics
Documents
File NameDescriptionSoftware TypeSoftware Version
Rejection.pdf Read-Only Document
Rejection.mcd Mathcad Computational Document Mathcad
Comments to: Kevin Lehmann at Lehman@princeton.edu
©Copyright Kevin Lehmann, 1997. All rights reserved. You are welcome to use this document in your own classes but commercial use is not allowed without the permission of the author.